Postdoctoral Researcher Positions in ICT2022-12-13T09:46:05+02:00
Apply now

Project Positions for Postdoctoral Researchers and Research Fellows in ICT (Helsinki, Finland)

Aalto University and the University of Helsinki are looking for Postdoctoral Researchers and Research Fellows in multiple areas of ICT, including: artificial intelligence and machine learning, data science, privacy and security, computational health, cybersecurity, and human-computer interaction. We are committed to fostering an inclusive environment with people from diverse backgrounds, and researchers from underrepresented groups are particularly encouraged to apply.

The deadline for applications is January 15th, 2023 at 11:59 PM (23:59 UTC+02:00). By applying to this call, organized by Helsinki Institute for Information Technology HIIT and Finnish Center for Artificial Intelligence FCAI, you use one application to apply to positions for both of our hosting institutions, Aalto University and the University of Helsinki. Aalto University and the University of Helsinki are the two leading universities in Finland in computer science and information technology. Both are located in the Helsinki Metropolitan area, and the employing university will be determined by the supervising professor. Please see below for additional information on the application process. (In the section entitled “How to Apply”)

The Helsinki Institute for Information Technology (HIIT) and the Finnish Center for Artificial Intelligence (FCAI) bring together the world-class expertise of Aalto University and the University of Helsinki in joint research centers and programs. There are multiple research groups from either Aalto University or the University of Helsinki participating in this open call. Recruited researchers will work in one of the participating groups, or in some cases, a shared position may also be possible.

Research Group Projects

These projects are not directly funded by HIIT, but by individual researchers or research groups in the Helsinki ICT community. Professors are either at Aalto University or the University of Helsinki. More information about the professors or the research groups can be found with their related project information below:

Project 1: Explainable and Robust AI for Scientists2022-12-13T08:51:53+02:00

Machine learning and AI are extensively used in the sciences. When modelling physical systems, the understandability and statistical robustness of the models is often more important than predictive accuracy. We are looking for talented postdoctoral researchers and doctoral students to study explainable and understandable AI and the uncertainty quantification of AI models. While the AI methods we develop are generic and not tied to any specific application domain, we work closely with scientists to build Virtual Laboratory for Molecular Level Atmospheric Transformations ( More information:

Keywords: explainable AI, uncertainty quantification

Focus Area(s): Artificial Intelligence, Data Science

Supervisor: Kai Puolamäki (Helsinki)

Project 2: Machine Learning for Affective Computing2022-12-13T08:52:52+02:00

We are seeking 1-2 postdoctoral researchers to conduct research in machine learning for affective computing and affective recommender systems. The postdocs will develop machine learning approaches for data captured from affective interfaces and brain-computer interfaces. The positions provide opportunities to contribute both to fundamental computer science and ground-breaking human-computer interaction / affective computing research. The positions are associated with the BANANA project ( and will be offered initially for a duration of 24-28 months depending on the starting date.

The ideal candidate has a strong background in applied machine learning and/or affective computing. However, a good understanding of machine learning methods and good programming skills are required.

Keywords: machine learning, affective computing, biomedical signals, eeg, fnirs, deep learning, recommender systems

Supervisor: Tuukka Ruotsalo (Helsinki)

Focus Area: Artificial Intelligence

Project 3: Green NLP – controlling the carbon footprint in sustainable language technology2022-12-13T08:53:48+02:00

GreenNLP addresses the problem of increasing energy consumption caused by modern solutions in natural language processing (NLP). Neural language models and machine translation require heavy computations to train and their size is constantly growing, which makes them expensive to deploy and run. In our project we will reduce the training costs and model sizes by clever optimizations of the underlying machine learning algorithms with techniques that make use of knowledge transfer and compression. Furthermore, we will focus on multilingual solutions that can serve many languages in a single model reducing the number of actively running systems. Finally, we will also openly document and freely distribute all our results to enable efficient reuse of ready-made components to further decrease the carbon footprint of modern language technology. The project will run in collaboration with the University of Turku and CSC IT center of science. The position is for 2 years with a possibility of an extension until the end of that project in December 2025.

Keywords: Language technology, NLP, deep learning, large language models, machine translation, knowledge distillation

Supervisor: Jörg Tiedemann (Helsinki) (in collaboration with TurkuNLP and CSC)

Focus Area: Artificial Intelligence

Project 4: Computational analysis and modeling of multi-level data in cancer research2022-12-13T08:54:45+02:00

The open position is in a research project funded by EU Horizon 2020 (DECIDER), which is one of the largest ovarian cancer precision oncology projects in the world. The goal of the DECIDER project is to develop mathematical models and data analysis methods based on genomics, clinical, histopathological and radiological data to characterize mechanisms driving drug resistance and suggest combinatorial treatments for treatment of relapsed cancer patients. Research is based on prospective, multi-region, longitudinal ovarian cancer patient clinical samples. A successful candidate has enthusiasm in science, strong background in systems theory, applied mathematics, machine learning, multivariate statistics, computer science or bioinformatics. Basic understanding of cancer biology, especially tumor evolution, is considered an asset. Applicants with strong signal processing expertise and interest in histopathological image analyses are encouraged to apply. Ability to work together with clinical/biological researchers is required.

Keywords: Modeling, statistics, cancer research, image analysis, tumor evolution

Supervisor: Professor Sampsa Hautaniemi (

Focus Area(s): Artificial Intelligence, Computational Health, Data Science

Project 5: Transformative AI Collaborations2022-12-16T12:27:53+02:00

Aalto University has launched a new collaborative initiative in transformative research to support cooperation between AI and Data Science researchers with researchers in other fields. The aim is to create new knowledge by developing AI methods to advance research in other domains.

We are especially seeking candidates interested in collaborations between Artificial Intelligence and/or Data Science with the following areas for Transformative AI collaborations:

  1. Automatic generation of production planning and scheduling optimization models from data: We aim at creating mathematical programming models by fitting a black box neural network model to the data collected from real or simulated processes and converting it into a “classical” (MILP, NLP) optimization model. The approach would have several benefits: 1) Semi-automation in model generation; 2) Automatic model calibration – represents reality, at least on average; 3) Ready-made general solvers can be used for solving
    Keywords related to the current approach include: Explainable NNs, metamodeling, gray-box modeling
    Collaborating supervisor: Esko Niemi, Digital Production Group
  2. AI and Digital Twins in Advanced Manufacturing: The physical manufacturing processes are supplemented by their Digital Twins. Digital Twin is a concept that includes all the analytical capability and the data generated by the physical process in real time.  It is like weather forecasting for optimal manufacturing processes.  For the most optimal process we want to add sensors and other data gathering concepts for optimization.  The most of manufacturing process optimization with Digital Twins is based on analytical understanding of the process.  However, the amount of data that will be available will probably be beyond the capability of analytical models and this calls for new AI-based models. More information about the group is available on the website.
    Collaborating supervisor: Jouni Partanen, Additive Manufacturing Technology and Applications Group
  3. Deep learning and process-based model fusion for improved understanding of the earth system:  Earth system models are key to understanding the ongoing global change and potential mitigation strategies. Enhancing the capacity of the models underlying the predictions about the future conditions of our planet is crucial to mitigate and adapt to ongoing global environmental change. Towards this goal, deep learning methods have recently been applied also in the field of earth system modeling, contesting, or even surpassing the performance of traditional process-based models. These two methodologies approach the modeling problems from different perspectives; deep learning and other machine learning methods are data-driven, limited by data availability, whereas process-based models are mainly limited by model flexibility and our understanding of the underlying processes. This project focuses on understanding how to combine the strengths of these two types of models. Potential avenues to explore include, among others, 1) deep-learned model ensembles where deep learning informs an optimal model ensemble based on local conditions, 2) ensembles consisting of both data-driven and process-based models, 3) hybrid
    models integrating different aspects of DL and process-based models, or 4) identifying conditions where the strengths of each model type, data-driven or process-based, could be exploited. The desired outcome of this project is methodological advancements on how deep learning improves earth system modeling and practical applications leveraging the strengths of deep learning and process-based modeling in environmental and earth system sciences. More information about the research group is available on their website.

    Collaborating supervisors: Matti Kummu, Olli Varis, Water and Development Research Group
  4. Multilevel optimization of autonomously interacting textiles: In the Multifunctional Materials Design group, we target a paradigm shift for smart soft materials into user-customizable textiles that autonomously interact with their environment and communicate through changes in color or shape. This will be done by coupling modular thermo- and photo-actuators to different textile constructions ranging from traditional fiber arts to industrial knitting and weaving techniques. Optimizing responses of different textile architectures composed out of actuating yarns opens an interesting question of how to predict the behavior of a dynamic textile combining multiple functionalities. We are especially interested in combining physical modelling of interacting/actuating networks to data science approaches in order to predict the dynamic behavior of textiles. We are a welcoming group of researchers that embraces transdisciplinary methods and diverse – often surprising contributions (Multifunctional Materials Design | Aalto University).
    Collaborating supervisor: Jaana Vapaavuori, Multifunctional Materials Design Group
  5. Physically interpretable AI for power-conversion applications: AI can be used for condition monitoring of power converters and electrical machines that are used, for example, in renewable energy generation. For these kind of applications, physically interpretable AI methods are of interest. Furthermore, it would be important to be able to generalize data and models of one device design to other types of device designs.
    More information about the research group is available on their website
    Collaborating supervisor: Marko Hinkkanen, Electric Drives Group
  6. Biomaterials: The School of Chemical Engineering and School of Science host jointly the LIBER Centre of Excellence ( with multiple research directions having potential for transformative research in collaboration with artificial intelligence. The focus is on processes leading to emergent materials structures, complex functions of materials, and dissipative nonequilibrium processes. Functions that are developed are for example self-organization, regeneration, adaptability, and self-repair. Other possible research directions include working with new imaging and data acquisition methods, and new computationally-aided design methods of new materials.
    Collaborating supervisor: Markus Linder, LIBER Centre of Excellence
  7. Computational Design: Computational design applies AI methods to automate and assist professional designers in their tasks. Until recently, the field was dominated by optimization and model-based methods, but with deep nets has shifted into data-driven approaches. We believe that the next revolution will build on user models (models of designers), which are needed to make more relevant and timely suggestions for the human designer. We are looking for researchers interested in studying possibilities emerging in latest AI research, such as using LLMs to enable natural language interaction with AI assistants in design, and training multi-purpose transformers and diffusion models for design tasks. We are also interested in methods for inferring designers’ preferences interactively and using this knowledge to propose completions to design tasks
    Supervisor: Antti Oulasvirta
  8. Integrating Human Electrophysiological Models and Data using Bayesian methods: Bayesian workflow offers methods to constrain models with experimental data while retaining prior knowledge from e.g., neuroanatomical studies. In this project, we will use state-of-the-art Bayesian workflow tools (Kallioinen et al. (2022), Siivola et al. (2021), Gelman et al. (2020), Piironen et al. (2020)) to constrain BNMs with MEG data. Implementation: We will use (i) Bayesian Optimisation-based Likelihood-Free Inference (LFI) methods to estimate model parameters from MEG data, (ii) Projection Predictive Inference to inform decisions on parameters to include in the model, and (iii) Power-scaling approaches to diagnose sensitivity of the parameter estimates to model changes.
    Supervisors: Aki Vehtari, Matias Palva
  9. AI forecasting methods for renewable energy: Forecasting variable renewable energy generation and its impact on day ahead power prices. Especially prediction of wind power generation in a few coming days and modelling how the wind variation will impact power balance and thus the market prices.
    Supervisors: Simo Särkkä, Matti Lehtonen
  10. Pattern discovery and structure discovery in sequential data: The project looks at sequential data (e.g., data from the electric grid), and combines structure discovery and search for rare phenomena.
    Supervisor: Heikki Mannila

If you are particularly interested in one or more of the Transformative AI projects listed above, please indicate which project(s) you would be interested working with in your cover letter.


Project 6: Trust-M: Designing Inclusive & Trustworthy Digital Public Services for Migrants in Finland2023-01-03T11:49:28+02:00

The Trust-M research project aims to improve the integration of migrants in Finland by devising hybrid and trustworthy digital services based on conversational AI. Finnish public services may not always be accessible, inclusive or trustworthy for all migrants. Improving such services can strengthen social cohesion, resilience of the labor market, and economic vibrance in Finnish Society. The project is a partnership between Aalto University, University of Helsinki, Tampere University, and City of Espoo, supported by the Academy of Finland’s Strategic Research Council (SRC) program in Security and Trust in the Age of Algorithms (SHIELD).

Project objectives include: (1) understanding how the socially and culturally constructed notions of trust, inclusion and equality are manifest in present-day digital public sector services, (2) devising alternatives for novel digital public sector services that could nurture trust and respect human rights, particularly considering migrant women, and (3) designing pilot versions of hybrid digital services based on conversational interaction, in conjunction with the City of Espoo.

We are seeking motivated Doctoral and Postdoc researchers to join the Trust-M team to conduct research and design of novel trustworthy conversational AI systems using multimodal voice-based interaction. Candidates must ideally have interests and expertise in at least 2-3 relevant areas including Human Computer Interaction (HCI), Natural Language Processing (NLP), conversational AI chatbots, speech/voice interaction, rapid prototyping, design research, user evaluation, and ethical/responsible AI. Evidence of prior work and publications in one or more of these areas is highly beneficial. Good interpersonal skills, collaborative research, conducting ethical research studies and participatory design with end users, and/or project coordination experience is helpful. Diverse international candidates with multi-lingual backgrounds are encouraged to apply.

You would join the CRAI-CIS research group in the Computer Science department at Aalto University. The transdisciplinary group explores the impact of technology in critical societal contexts, working at the intersection of computational and social sciences engaging HCI and participatory design. More here:

Keywords: Human Computer Interaction (HCI), Natural Language Processing (NLP), conversational AI, speech/voice interaction, design research, human-AI interaction

Supervisor(s): Nitin Sawhney ( and Tom Bäckström (

Project 7: Bayesian Workflows for Iterative Model Building and Networks of Models2022-12-13T08:55:02+02:00

Statistical analysis is critical when it comes to obtaining insights from data. Despite the practical success of iterative Bayesian statistical model building, it has been criticized to violate pure Bayesian theory and that we may end up with a different model had the data come out differently. In this project, we formalize and develop theory and diagnostics for iterative Bayesian model building. The practical workflow recommendations and diagnostics guide the modeller through the appropriate steps to ensure safe iterative model building, or indicate when the modeler is likely to be in the danger zone.

Sometimes the user wants also to build and compare models of different complexity or based on different assumptions.  Similar workflow ideas can be used to support also analysis of networks of models, making it easier to illustrate similarities and important differences. By making applied scientific research and data analysis more reliable and reproducible, our understanding of the world and decision-making will be improved. Related paper and video

Supervisor: Aki Vehtari (Aalto)

Focus Area: Artificial Intelligence

Project 8: Uncertainty-aware and explainable deep learning for medical data2022-12-13T08:55:35+02:00

Deep learning has achieved remarkable performance in various medical classification, regression, and semantic segmentation tasks. However, the current approaches rarely estimate uncertainty in the predictions, which translates to improper level of trust for the models. In this project, the post-doc researcher will contribute to the development and application of uncertainty-aware (such as approximate Bayesian) deep learning methods on various data domains. In addition, the explainability and controllability of deep learning models is of interest to further improve the trustworthiness of these systems. We have the following ongoing interdisciplinary collaborations:

  • Helsinki University Hospital (HUS) and Central Finland Central Hospital for clinical diabetic retinopathy classification using fundus images and additional metadata.
  • Tampere University Hospital (TAUH) and Tampere University (TAU) for various maxillofacial localisation and segmentation tasks from cone-beam computed tomography images.
  • MD Anderson Cancer Center (The University of Texas), HUS and TAUH for various tasks for patients with head and neck cancer or sarcopenia such as treatment planning, surgical planning and survival analysis from multimodal imaging data (MRI/CT/PET).

A successful applicant is expected to have a degree in machine learning or related field, have experience in modern deep learning libraries (e.g. Pytorch), and have the capacity to develop and apply state-of-the-art models, and to handle real world data.

Supervisor: Kimmo Kaski (Aalto)

Keywords: Deep learning, Uncertainty, Medical data, Medical imaging, Explainability, Classification, Segmentation

Focus Area(s): Artificial Intelligence, Computational Health

Project 9: Machine Learning for Health (ML4H)2022-12-13T08:56:10+02:00

Accumulation of massive amounts of health data has enabled researchers to address questions such as: how to accurately predict the risk of disease, how to personalize treatments based on real-time data from wearable devices, or how to use genomic data to understand disease or antibiotic resistance. Central challenges in ML4H include noisy data, multiple heterogeneous data sources, learning about causality, interpreting the models, and quantifying the uncertainty, to name a few. We tackle these by developing models and algorithms which leverage modern machine learning principles: Bayesian machine learning, deep latent variable models, Gaussian processes, transformers, reinforcement learning, and natural language processing. Successful applicants are expected to have outstanding skills in machine learning, statistics, applied mathematics, or a related field. The goal of the position is to develop novel methods for challenging biomedical applications. Examples of our recent research can be found in

Supervisor: Prof. Pekka Marttinen (

Focus Area(s): Artificial Intelligence, Computational Health, Data Science

Project 10: Statistical Genetics and Machine Learning2022-12-13T08:57:10+02:00

The emergence of genomic sequencing and other molecular profiling methods together with the digitalization of health care data collections have created unprecedented opportunities for better treatment options. INTERVENE is an international and interdisciplinary consortium that seeks to leverage these vast but underused data resources to generate clinically actionable knowledge for improved understanding of diseases and treatment options tailored to individuals, by analyzing altogether 1.7 million sequenced genomes with longitudinal clinical data. We are looking for an outstanding  postdoctoral researcher to join the interdisciplinary and international group of researchers in INTERVENE. From an applicant we expect a PhD in statistical genetics or a related field, or a PhD in a methodological field (e.g., machine learning, applied mathematics, statistics) coupled with a strong interest to apply the skills in the genetics application. More information about the project:

Supervisor: Prof. Pekka Marttinen (

Focus Area: Artificial Intelligence, Computational Health, Data Science

Project 11: Machine learning for wireless and mobile systems2022-12-13T08:57:33+02:00

Machine learning (ML) is the foundation of artificial intelligence in today’s applications. The scope of ML is wide, including (but not limited to) speech synthesis and recognition, machine translation, and computer vision. This project involves designing and (or) applying ML techniques to the scenarios represented by wireless and mobile systems.

The related research encompasses: design of efficient deep neural network architectures for embedded devices; protocols and algorithms for distributed inference; lifelong learning for optimization of wireless systems; applications of ML to virtual and augmented reality. The project requires strong analytical skills, proficiency in using ML frameworks such as Tensorflow, and keen interest in applications of ML.

Supervisor: Mario Di Francesco (Aalto)

Keywords: machine learning, wireless communication, mobile computing, embedded systems, augmented reality, optimization, lifelong learning

Focus Area: Artificial Intelligence

Project 12: Cloud-native systems: software systems and security2022-12-12T11:19:33+02:00

Cloud computing has introduced a new model for provisioning resources and services over the Internet on top of a virtualized, elastic infrastructure. Additional paradigms have also recently emerged: edge, fog, and serverless. Moreover, modern (namely, cloud-native) applications are generally composed of multiple microservices, realized as software containers and managed through an orchestrator such as Kubernetes. This project involves addressing different aspects of cloud-native systems, with focus on software systems and security. The related research encompasses: design and analysis of techniques for scalable and reliable computing; applying network economics to different computing paradigms and their applications; addressing open challenges in cloud network security; characterizing the performance and the security of applications based on microservices. The project requires strong analytical skills, proficiency in programming distributed systems, and solid knowledge on cloud-native software, in particular, Kubernetes.

Supervisor: Mario Di Francesco

Keywords: cloud computing, edge computing, serverless, microservices, cloud security, network economics, distributed systems.

Project 13: Neuro-symbolic multiagent reinforcement learning2022-12-13T08:58:10+02:00

Researchers interested in postdoctoral/doctoral positions are invited to apply for this project. Here, we shall study a new approach to synthesis of efficient communication schemes – including learning of novel concepts – in cooperative multi-agent systems, trained via reinforcement learning. We combine symbolic methods with machine learning, in what is referred to as a neuro-symbolic system, where a neural network learns to produce programs in a symbolic language to solve a task at hand.  Agents will not be restricted to use only primitives provided as input, but interactions will be interleaved with (symbolic) steps to extend and revise their current language with novel higher-level concepts, allowing for generalisation and more informative messages. We believe this combination of neural and symbolic methods will be an important next step in the development of AI beyond today’s capabilities, including bridging the gap between black-box neural models and classical symbolic methods. See [1] for a gentle introduction to the area.

The postdoc funding is assured for two years, while the PhD funding is for four years. Research for this project will be conducted at Aalto University, Finland in a world-class group known for its contributions to representation learning, generative models, and multiagent systems [2-8]

The selected candidate will work closely with the group of Prof. Moa Johansson (Chalmers University, Sweden).  There might also be an opportunity to work with our collaborators at MIT.

Contact: Vikas Garg (  or


[1] E. Jergéus, L. Karlsson Oinonen, E. Carlsson and M. Johansson. Towards Learning Abstractions via Reinforcement Learning.  8th Int. Workshop on Artificial Intelligence and Cognitive Science (2022).

[2] V. Garg and T. Jaakkola.  Predicting deliberative outcomes. ICML (2020).

[3] V. Garg, S. Jegelka, and T. Jaakkola. Generalization and representational limits of graph neural networks. ICML (2020).

[4] Y. Verma, S. Kaski, M. Heinonen, and V. Garg. Modular Flows: Differential Molecular Generation. NeurIPS (2022).

[5] A. Souza, D. Mesquita, S. Kaski, and V. Garg. Provably expressive temporal graph networks. NeurIPS (2022).

[6] J. Ingraham, V. Garg, R. Barzilay, and T. Jaakkola. Generative Models for Graph-Based Protein Design. NeurIPS (2019).

[7] V. Garg and T. Jaakkola. Local Aggregative Games. NeurIPS (2017).

[8] V. Garg, T. S. Jayram, and B. Narayanaswamy. Online optimization with dynamic temporal uncertainty, AAAI (2013).


Supervisor: Vikas Garg (Aalto)

Keywords: Reinforcement Learning, Multiagent systems, Graph Neural Networks, Symbolic methods, Program Synthesis, Game Theory

Focus Area: Artificial Intelligence

Project 14: Transformer methods for hyperspectral image fusion2022-12-13T08:58:22+02:00

The project will study and develop new Transformer-based models for modeling of multispectral, hyperspectral and lidar based remote sensing data in varying resolutions. We will analyze the information content of modern and future hyperspectral earth observation and lidar data obtained over the boreal forest zone. The results will be used for studying the biodiversity of forests based on the distrubution of tree species and the spatial arrangements of individual trees. In addition, the models will be included in a digital twin of the Finnish forests. The project aims at novel methodological and algorithmic improvements by using state-of-the-art Transformer models that have already been studied and developed in the research group.

Supervisor:  Jorma Laaksonen, Senior University Lecturer (Aalto)

Keywords: machine learning, deep learning, transformers, computer vision, data fusion, remote sensing, hyperspectral data

Focus Area: Artificial Intelligence

Project 15: Machine learning for drug design2022-12-13T09:37:57+02:00

Recent progress in machine learning for generative and predictive models of molecules brings us towards computational, automatised drug design. We develop statistical methods and models for molecular structures, energies and interactions with the help of deep learning. A number of open problems reside in developing neural network models with physics-based inductive biases, in generative models in 3D spaces, in modelling the property landscapes of molecules, and in generalizing outside the training distribution in molecular design.

We are looking for motivated candidates with background in computational sciences, machine learning, statistics. You will join the Probabilistic Machine Learning (PML) research group at Aalto,

Supervisor(s): Samuel Kaski, Markus Heinonen

Keywords: probabilistic modelling, drug design, deep learning

Focus Area: Artificial Intelligence

Project 16: Probabilistic modelling and Bayesian inference for machine learning2022-12-13T09:38:09+02:00

I am looking for a new postdoc in my team which develops probabilistic modelling and Bayesian inference methods. The team has several exciting new machine learning formulations we work on, and opportunities for applying the methods with top-notch collaborators. But the core is always development of new methods, and with this call I am looking for talented researchers with background in machine learning, stats or CS (or other directly relevant topics) who are keen on developing the new methods. In the cover letter, let me know what you are interested in – if we are already working on it, all the better, but I am willing to listen to new ideas too. Keywords include: probabilistic modelling, Bayesian inference, simulator-based / likelihood-free inference, multi-agent RL and collaborative AI, sequential decision making and experimental design, privacy-preserving inference, Bayesian deep learning. Part of the team is in Finland/EU and part in the UK where I run a Turing AI World-Leading Researcher project. Plenty of collaboration opportunities exist between those, in ELLIS, and overseas.

Supervision: Samuel Kaski

Focus Area: Artificial Intelligence

Project 17: Multi-agent modeling for human-AI interaction2022-12-13T09:38:48+02:00

We develop AI assistants which help people make better decisions, with ongoing applications in science and engineering. We use multi-agent formalisms to define the assistance problems these assistants solve, including the human being assisted, and employ models of human behavior to (pre-)train them in silico. We build models of human behavior using (multi-agent) reinforcement learning, based on theories of human behavior from cognitive science.

In this project you will develop novel multi-agent formalizations of assistance and create new models of human behavior for these formalizations. The focus will be on maximally autonomous assistants — assistants that automate a person’s task as much as possible, while using a minimal interaction to learn to solve said person’s task well.

We are looking for a postdoc and doctoral student with experience in probabilistic machine learning and reinforcement learning. No formal experience with cognitive science is required. Additional knowledge in any of the following will be helpful: game theory, multi-agent RL, Bayesian RL, computational rationality, and inverse reinforcement learning.
This project will involve collaborations with Prof. Ville Kyrki (robotics), Prof. Andrew Howes (cognitive science), TU Delft, MIT, and pharma and self-driving car companies.

Supervisor(s): Samuel Kaski (Aalto),  Andew Howes (Birmingham)

Keywords: user modeling, multi-agent RL, human-AI interaction, cooperative AI

Focus Area: Artificial Intelligence

Project 18: Maximally autonomous AI2022-12-13T09:39:05+02:00

Current machine learning approaches have shown outstanding success in various tasks. However, they generally require an explicitly defined goal, for instance as a reward or objective function. Defining these goals in real-world applications is laborious and error-prone, often leading to misaligned and undesirable behavior. We develop assistants that infer people’s goals through interaction, and thus avoid the need for well-specified goals.

In this project, you will develop principles and methods for AI assistants to be maximally autonomous. The goal is to automate to the greatest extent possible, while using a minimal interaction with the person being assisted. Interaction is important to learn the person’s goal, but should be done sparingly. It should only happen when necessary, i.e. when it can help reduce uncertainty about the goal and improve the assistant’s long-term decision making.

We are looking for a postdoc and doctoral student with experience in probabilistic machine learning and reinforcement learning. No formal experience with cognitive science is required. Additional knowledge in any of the following will be helpful: game theory, multi-agent RL, Bayesian RL, computational rationality, and inverse reinforcement learning.
This project will involve collaborations with Prof. Ville Kyrki (robotics), Prof. Andrew Howes (cognitive science), TU Delft, MIT, and pharma and self-driving car companies.

Supervision: Samuel Kaski (Aalto), Ville Kyrki (Aalto)
Keywords: user modeling, multi-agent RL, human-AI interaction, cooperative AI

Focus Area: Artificial Intelligence

Project 19: Deep learning with differential privacy2022-12-13T09:39:16+02:00

Differential privacy allows developing machine learning algorithms with strong privacy guarantees. Recent work shows it is possible to combine strong privacy and high accuracy by pre-training models on public data and only fine-tuning the model with the sensitive data. However, high accuracy still requires care for example in hyperparameter tuning. The aim of this project is to develop methods that make it easier to train high accuracy private models. The project will benefit from a very large grant of compute time on LUMI, 3rd fastest supercomputer in the world. The project requires a background in deep learning.

Supervisior(s): Antti Honkela (Helsinki), Samuel Kaski (Aalto)

Keywords: Deep learning, image classification, hyperparameter learning, differential privacy

Focus Area: Artificial Intelligence

FCAI positions for Winter 2023 Call

Finnish Center for Artificial Intelligence FCAI is an international research hub initiated by Aalto University, the University of Helsinki, and the Technical Research Centre of Finland VTT. We are a part of the pan-European ELLIS AI network – we host ELLIS Unit Helsinki and coordinate the European network of AI excellence centers ELISE.We are looking for postdocs and PhD students to join our community of machine learning researchers. Our positions are in the areas of reinforcement learning, probabilistic methods, simulator-based inference, privacy and federated learning, and multi-agent learning. For more information, please see FCAI webpage:

Project F1: AI-assisted design2022-12-12T15:12:24+02:00

FCAI is working on a new paradigm of AI-assisted design that aims to cooperate with designers by supporting and leveraging the creativity and problem-solving of designers. The challenge for such AI is how to infer designers’ goals and then help them without being needlessly disruptive. We use generative user models to reason about designers’ goals, reasoning, and capabilities. In this call, FCAI is looking for a postdoctoral scholar or research fellow to join our effort to develop AI-assisted design. Suitable backgrounds include deep reinforcement learning, Bayesian inference, cooperative AI, computational cognitive modelling, and user modelling.

Example publications by the team

Keywords: AI-assisted design, user modeling, cooperative AI

Supervisor(s): Profs. Antti Oulasvirta (Aalto University), Samuel Kaski (Aalto University), Perttu Hämäläinen (Aalto University)


Project F2: Amortized inference for experimental design and decision making2022-12-12T15:16:16+02:00

We develop amortized experimental design and inference techniques that take into account the down-the-line decision making task. For example, this may include delayed-reward decision making where data has to be measured, at a cost, before making the decision. This problem occurs in the design-build-test-learn cycles which are ubiquitous in engineering system design, and experimental design in sciences and medicine. The solutions need Bayesian experimental design techniques able to work well with simulators, measurement data and humans in the loop, who are both information sources and the final decision makers. For online and real-time tasks, algorithmic recommendations need to come near-instantly, thus requiring amortization of both experimental design and of the decision-making suggestions. The assistive methods need to account for uncertainty in the inference process and possibly in the utility function itself.

We are looking for a machine learning researcher with familiarity with probabilistic modelling, amortized inference via deep learning techniques, and/or Bayesian experimental design, interested in developing the new methods, with options on applying the techniques to improve modelling in the FCAI’s Virtual Laboratories.

Keywords: Sequential design of experiments, Bayesian experimental design, active learning, amortized inference

Supervision: Luigi Acerbi (University of Helsinki), Samuel Kaski (Aalto University)

Project F3: Amortized surrogates for simulation and inference2022-12-12T15:18:50+02:00

Recent advances in machine learning have shown how powerful emulators and surrogate models can be trained to drastically reduce the costs of simulation, optimization and Bayesian inference, with many trailblazing applications in the sciences. In this project, the candidate will join an active area of research within several FCAI groups to develop new methods for simulation, optimization and inference that combine state-of-the-art deep learning and surrogate-based kernel approaches – including for example deep sets and transformers; normalizing flows; Gaussian and neural processes – with the goal of achieving maximal sample-efficiency (in terms of number of required model evaluations or simulations) and wall-clock speed at runtime (via amortization). The candidate will apply these methods to challenging problems involving statistical and simulator-based models that push the current state-of-the-art, be it for number of parameters (high-dimensional amortized inference), number of available model evaluations (extreme sample-efficiency) or amount of data. The ideal candidate has expertise in both deep learning and probabilistic methods (e.g., Gaussian processes, Bayesian optimization, normalizing flows).


Keywords: Emulators, amortized inference, Bayesian optimization, normalizing flows, simulator-based inference

Supervisior(s): Profs. Luigi Acerbi (University of Helsinki), Jukka Corander (University of Helsinki)

Project F4: Automatic model quality assessment and improvement2022-12-12T15:22:20+02:00

The objective of this research topic is to enable autonomous embodied systems to automatically assess and improve their internal models, which guide their interaction with the external world. The task is to develop application independent methods that automatically evaluate the quality of learned models and propose and execute the model improvement. The developed methods will be included in the broader work on autonomous driving.

The proposed research activity takes a step toward the genuinely long-life operation of autonomous agents. The work will focus on developing methodologies that provide information about the correctness of an entire model or its parts. This information will later be utilized in the iterative or online learning process such that the model will be selectively updated. Thus there will be no need to retrain the entire model from scratch, and there will be no risk of decreasing the overall performance. In this work, the candidate will also look into the problem of explainability so that motivation for the proposed model changes can be provided.

The candidate should have a PhD in Computer Science, Machine Learning, AI, Robotics or related fields. They should have an excellent track record in machine learning or active perception; experiences with explainable AI or robotic introspection are a plus.

Keywords: Model quality assessment, model introspection, long-term autonomy

Supervision: Profs. Tomasz Kucner (Aalto University); Joni Pajarinen (Aalto University)

Project F5: Computational rationality2022-12-12T15:24:32+02:00

Computational rationality is an emerging integrative theory of intelligence in humans and machines [1] with applications in human-computer interaction, cooperative AI, and robotics. The theory assumes that observable human behavior is generated by cognitive mechanisms that are adapted to the structure of not only the environment but also the mind and brain itself [2]. Implementations use deep reinforcement learning to approximate optimal policy within assumptions about cognitive architecture and their bounds. Cooperative AI systems can utilize such models to infer causes behind observable behavior and plan actions and interventions in settings like semiautonomous vehicles, game-level testing, AI-assisted design etc. FCAI researchers are at the forefront in developing computational rationality as a generative model of human behavior in interactive tasks (e.g., [3,4,5]) as well as suitable inference mechanisms [5]. We collaborate with University of Birmingham (Prof. Andrew Howes) and Université Pierre et Marie Curie (UPMC, CNRS) (Dr. Julien Gori, Dr. Gilles Bailly).

In this call, we are looking for a talented postdoctoral scholar or research fellow to join our effort to develop computational theory as a model of human behavior. Suitable backgrounds include deep reinforcement learning, computational cognitive modelling, and reinforcement learning.


  1. S. Gershman et al. Computational rationality: A converging paradigm for intelligence in brains, minds, and machines. Science 2015.
  2. R. Lewis, A. Howes, S. Singh. Computational rationality: A converging paradigm for intelligence in brains, minds, and machines. Topics in Cognitive Science 2014.
  3. J. Jokinen et al. Parameter Inference for Computational Cognitive Models with Approximate Bayesian Computation. Proc. CHI’21, ACM Press.
  4. C. Gebhardt et al. Hierarchical Reinforcement Learning Explains Task Interleaving Behavior. Computational Brain & Behavior 2021.
  5. J. Takatalo et al. Predicting Game Difficulty and Churn Without Players. Proc. CHI Play 2020.
  6. A. Kangasrääsiö et al. Parameter Inference for Computational Cognitive Models with Approximate Bayesian Computation. Cognitive Science 2019.

Supervisior(s): Profs. Antti Oulasvirta (Aalto University), Andrew Howes (University of Birmingham), Samuel Kaski (Aalto University), Arto Klami (University of Helsinki), Perttu Hämäläinen (Aalto University)

Keywords: Computational rationality, computational cognitive modeling, deep reinforcement learning

Project F6: Deep learning for material science2022-12-12T15:31:18+02:00

The goal of the project is to develop novel deep learning algorithms to answer open questions in nanoscale physics such as predicting molecular structure in liquids or developing accurate, but very fast models of water. In particular we want to take advantage of the capabilities of Graph Neural Networks to provide robust and highly efficient simulators for molecular systems. These models will be coupled to state-of-the-art experimental characterisation, ultimately including a dynamic interaction where simulations are actively used to focus on information rich regions during experiments. We are looking for applicants with a strong background in deep learning and/or physical simulations.

Keywords: Deep learning, graph neural networks, material science, physics simulations

Supervisior(s): Profs. Adam Foster (Aalto University), Alexander Ilin (Aalto University)

Project F7: Deployment as a fundamental ML challenge2022-12-12T15:31:34+02:00

Machine learning (ML) is now used widely in sciences and engineering, in prediction, emulation, and optimization tasks. Contrary to what we would like to think, it does not work well in practice. Why?

Because the conditions during deployment may radically differ from training conditions. This has been conceptualized as distribution shift or sim-to-real gap, and a particularly interesting challenge which we will be tackling is changes due to unobserved confounders. Solving this challenge is imperative for widespread deployment of ML. In this project, we will consider deployment as a fundamental machine learning challenge, developing new principles and methods for tackling this problem which can be argued to be the main show-stopper in making machine learning seriously useful in solving the real problems we are facing, in sciences, companies and society. We have exciting test cases in FCAI’s highlight problems, in drug design, materials science, and health applications. We are looking for candidates with strong background in probabilistic machine learning.

Keywords: ML deployment, distribution shift, out-of-distribution, unobserved confounders

Supervision: Profs. Samuel Kaski (Aalto University), Vikas Garg (Aalto University)

Project F8: Efficient probabilistic modeling of speech2022-12-13T08:06:44+02:00

Probabilistic models are essential for capturing the stochastic properties in speech and audio signals. Speech synthesis in particular has recently emerged as a proving ground application for deep generative models. Autoregressive (AR) density models, such as WaveNet, achieve high quality, can be trained effectively using maximum likelihood, and have low algorithmic latency suitable for real-time applications. However, available implementations of AR inference using GPUs and popular Python libraries are slow, which has shifted research to feedforward models for parallel inference. These models can similarly achieve high quality, whether trained as Generative Adversarial Networks (GANs), Diffusion Models or Energy Based Models (with contrastive estimation). However, these training recipes often involve a complex mixture of training objectives, while feedforward models can further be relatively inefficient and introduce algorithmic latency. Meanwhile, Digital Signal Processing (DSP) methods for speech and audio contain valuable knowledge both in perceptually relevant metrics for similarity, and in design and implementation of models with feedback. First, we propose to leverage this knowledge to develop efficient DSP-based recurrent building blocks (with forward and backward stability guarantees), and integrate them into a deep learning system. Second, we aim to formulate a unified framework for evaluating the various probabilistic models for speech using Score Matching Networks, experiment on what is really needed to make a model work, and extend the current approaches to use perceptually relevant score matching functions. We are looking for applicants with a strong background in deep learning and probabilistic modeling.

Keywords: Deep generative models, digital signal processing, probabilistic modeling, speech

Supervision: Profs. Lauri Juvela (Aalto University), Alexander Ilin (Aalto)

Project F9: Evaluating and improving posterior inference for difficult posteriors2022-12-12T15:38:36+02:00

Both MCMC and distributional approximation algorithms (variational and Laplace approximations) often struggle to handle complex posteriors, but we lack good tools for understanding how and why. We study diagnostics for identifying the specific nature of the computational difficulty, seeking to identify e.g. whether the difficulty is caused by narrow funnels or strong curvature. We also develop improved inference algorithms that account for these challenges, e.g. via automated and semi-automated transformations for making the posterior easier or by better accounting for the underlying geometry. We are looking for applicants with a strong inference background and interest in working on improving inference for the hardest problems.

Keywords: MCMC, variational approximation, differential geometry, inference diagnostics, Bayesian workflow

Supervisor(s): Profs. Aki Vehtari (Aalto University) and Arto Klami (University of Helsinki); primary/secondary depending on the candidate’s interests

Project F10: Explainable AI for virtual laboratories2022-12-12T15:40:31+02:00

FCAI is actively developing methods and software for virtual laboratories to enable AI assistance in the research process. We are looking for a candidate to research explainable AI and uncertainty quantification. Efficient human-AI collaboration requires methods that are either inherently capable of providing explanations for the decisions or can explain the decisions of other AI models. For instance, the user needs to know why AI recommends a particular experiment or predicts a specific outcome. They should always be aware of the reliability of the AI models. You will conduct the project with a team of AI researchers with access to researchers specialised in various application areas. The applicant should be interested in incorporating the techniques as part of virtual laboratory software developed at FCAI for broad applicability.

Keywords: Virtual laboratory, explainable AI, uncertainty quantification, human-AI collaboration

Supervision: Profs. Arto Klami (University of Helsinki), Kai Puolamäki (University of Helsinki)

Project F11: Foundation models for interactive computing2022-12-12T15:44:41+02:00

Foundation models such as GPT-3 and CLIP are revolutionizing how AI is developed and applied, by providing reusable and general-purpose building blocks with unprecedented capabilities. Instead of training large-scale models from scratch for thousands or millions of GPU hours, one can solve novel tasks by combining pretrained foundation models in novel ways, or finetuning them for downstream tasks. The release of OpenAI’s CLIP, for instance, soon led to the emergence of CLIP-guided image generation models such as Disco Diffusion, Stable Diffusion, Midjourney, and DALL-E 2, as well as smaller-scale experiments such as ClipDraw and StyleClipDraw. CLIP also increasingly empowers various semantic search solutions across multiple industries.

The objective of this research topic is to study, develop, and test/validate foundation models for interactive computing, where they have received relatively less attention so far. The specific research foci depend on the hired researchers’ interests, but might include:

  • Large language models (LLMs) as textual human simulacra, e.g., in generating synthetic user research data, role-playing research participants in rapid research exploration and piloting.
  • Foundation models for simulated human movement control, such as AI “user simulators” that can be used to test interactive systems, or virtual AI motion capture actors that can follow choreographer instructions to generate animations or suggest ways to solve movement problems.
  • Multimodal extensions of the above, e.g., video-language reward models that determine how well an AI agent solves a task defined using natural language, based on visual observations, or user motivation and emotion models that can provide synthetic “think aloud” narrative based on user simulation data.
  • Generative models for designing interactive systems, e.g., generating visual designs, personas, user stories, or wireframes.

Aalto FCAI teams have already made progress in many of the directions above (e.g.,,,,, providing an excellent foundation for future research.

Keywords: Large Language Models, Foundation Models, User simulation, Generative models

Supervisor(s): Profs. Perttu Hämäläinen (primary, Aalto University); Robin Welsch (Aalto University), Christian Guckelsberger (Aalto University), Jaakko Lehtinen (Aalto University, NVIDIA)

Project F12: Interactive AI using multimodal communication2022-12-12T15:46:12+02:00

Intelligent machines require not only an internal model of the world but also interaction with humans and their environment. They need to make use of contextualized information and must be able to adapt to the user. This project focuses on multimodal communication that is natural for humans but difficult for machines and will deal with spontaneous spoken and written language, gestures and other forms of non-verbal types of interaction. The important aspect is to model those channels in combination to be able to deal with complementary information coming from audio-visual signals as well as traditional types of input coming from keyboards and touch-sensitive devices. We envision the development of a multimodal assistant that can interact with its users in a coherent natural way. The project will include work on cross-modal attention and sequence models that are able to capture long-distance dependencies and we will test our ideas in applications related to health and wellbeing. The framework for this work will be based on modern neural architectures and deep learning and combines aspects of supervised, unsupervised and reinforcement learning. This research problem will be solved as a collaboration between three research groups: Aalto ASR, Aalto Video Content Analysis and Helsinki-NLP, all in FCAI.

Keywords: Multimodal NLP, speech technology, computer vision

Supervision: Profs. Mikko Kurimo (primary, Aalto University), Jorma Laaksonen (Aalto University), Jörg Tiedemann (University of Helsinki)

Project F13: Language-empowered world models for RL2022-12-12T15:48:19+02:00

Incorporating accurate causal knowledge about an environment in terms of objects and their relationships into the world model of a reinforcement learning agent can yield significant improvements in reducing the amount of exploration required to solve new tasks and to generalize to new environments. Recently, causal representation learning has been proposed as a way of extracting and representing such causal knowledge from previous experience in the latent space. However, in practice data are extremely correlated and it is not straightforward to come up with ways to even disentangle relevant objects, not to mention the causal relationships between them. On the other hand, large language models (LLMs) encompass a lot of information about objects and causal relationships, for example: “User: How can I turn the lights on? GPT-3: Depending on the type of light fixture in the room, you can usually turn the lights on by flipping a switch near the entrance to the room.” Leveraging this type of information when learning world models for reinforcement learning (RL) and causal inference is still mostly underexploited, although the combination of LLMs with RL is a rapidly surfacing new topic in the machine learning community, see We are looking for a postdoc who would leverage and expand our previous work on RL agents with language understanding abilities, to address this highly topical research theme. In particular, we focus on improving the s-o-t-a on language guided RL benchmarks by empowering the agent with deep latent variable based probabilistic world models, which accurately encompass uncertainty that serves as the foundation for reliable planning.

Keywords: Causality, language modeling, reinforcement learning, world models

Supervision: Pekka Marttinen (Aalto); Alexander Ilin (Aalto)

Project F14: Long term planning with search graphs2022-12-12T15:49:42+02:00

In many complex sequential decision making tasks, online planning is crucial for high-performance. Monte Carlo Tree Search (MCTS) is an efficient online planning tool which employs a principled mechanism for trading off between exploration and exploitation. Following the success of MCTS in discrete control problems (such as the games of Go, Chess, and Shogi), various MCTS extensions have been proposed to continuous domains. However, the inherent high branching factor and the resulting explosion of the search tree size is limiting existing methods. In this project, we investigate novel extensions of MCTS for continuous domains based on search graphs. Our approach is built on the idea that sharing the same action policy between several states can yield efficient planning and thus high performance. This results in a limited number of stochastic action bandit nodes to produce a layered graph instead of an MCTS search tree allowing for long term planning. The designed algorithms can be used for robotic manipulation and navigation, for example, with a Boston Dynamics Spot robot. We are looking for applicants with a strong background in reinforcement learning (especially model-based) and tree search algorithms.

Supervision: Profs. Joni Pajarinen (Aalto University), Alexander Ilin (Aalto University)

Keywords: Continuous control, graph search, Monte Carlo tree search, online planning, reinforcement learning

Project F15: ML4Science2022-12-12T15:51:27+02:00

Machine learning is increasingly being used as a key element in research, for instance to efficiently approximate computationally costly simulations, automate design of experiments, and for integrated analysis of experimental results and multi-fidelity simulations. Much of the practical work is done in the context of specific applications in science, but our interest lies in the more general question of how ML could be used as part of the research process, essentially to improve the results and the scientific process itself. We seek solutions that work across multiple disciplines and applications. We are looking for candidates interested in aspects such as (a) how to best incorporate domain knowledge into probabilistic ML models, (b) how to integrate ML models as a part of the research process that involves e.g. also empirical experimentation, and (c) how to assist the research process itself using AI solutions. The work relates closely to our Virtual Laboratories initiative: realizing that most fields now use computational tools, essentially doing experiments first virtually with simulations, we can have scale advantages with AI methods that are cross-usable across the fields. The virtual laboratories give opportunities to demonstrate and validate the research contributions in several different natural science applications. An ideal candidate has expertise in both ML and some science domain, but candidates with strong background in either one are also considered.

Keywords: Natural science, virtual laboratory, AI-assisted research

Supervision: Profs. Samuel Kaski (Aalto University), Arto Klami (University of Helsinki), other professors (e.g. Patrick Rinke, Aalto University) depending on candidate’s interests/qualifications

Project F16: Multi-agent RL for collaborative AI2022-12-12T15:52:51+02:00

We develop the new ML principles and methods needed by AI assistants to help people make better decisions, with ongoing applications in science and engineering. We use multi-agent formalisms to define the assistance problems these assistants solve, including the human agent being assisted, and develop new multi-agent RL solutions for the problem. We are particularly interested in (1) how to build and (pre-)train models of human behaviour based on cognitive science, and (2) how to solve new ad-hoc teamwork problems with multi-agent RL.

We are looking for new members in our team, with experience in probabilistic machine learning and multi-agent reinforcement learning. No formal experience with cognitive science is required. Additional knowledge in any of the following will be helpful but not necessary – we have a great team to work with: game theory, Bayesian RL, computational rationality, and inverse reinforcement learning.

Recent publications by the team:

  1. (best paper award; HiLL@Neurips-22)
  2. (AAAI-23)
  3. (AAMAS-22)

Keywords: User modeling, multi-agent RL, human-AI interaction, cooperative AI

Supervision: Samuel Kaski (Aalto University) and other professors

Project F17: Private federated learning2022-12-12T15:54:13+02:00

Many applications of machine learning require training on distributed data while keeping the data private. Private federated learning enables this, but its communication requirements can be impractical. The aim of this project is to develop new approaches and methods for private learning on distributed data. A strong background in differential privacy and/or federated learning is an asset for this project.

Keywords: Differential privacy, federated learning, deep learning

Supervision: Profs. Antti Honkela (University of Helsinki), Samuel Kaski (Aalto University)

Project F18: Sustainability in computing2022-12-12T15:55:25+02:00

Sustainability is important in the context of AI and in particular there exist AI/ML/DS methods to improve sustainability of a given system using say AI methods. However, the computational methods used in AI are not always sustainable as they might require a lot of data or a lot of computational power produce the results. The aim of this research project is to take a look at sustainable AI methods and in particular sustainable computational methods whose energy fingerprint is minimal. One possible approach that has recently been studied is the clever use of parallel and distributed algorithms to decrease the amount of energy per flop. We are looking for applicants with interests in energy efficient or otherwise sustainable computational methods in general.

Keywords: Sustainability in AI, energy-efficient machine learning, data-light computing, parallel computing

Supervision: Profs. Simo Särkkä (Aalto University), Laura Ruotsalainen (University of Helsinki)

Project F19: Workflows for better priors2022-12-12T15:56:41+02:00

Bayesian models rely on prior distributions that encode knowledge about the problem, but specifying good priors is often difficult in practice. We are working on multiple fronts on making it easier, with contributions to e.g. prior elicitation, prior diagnostics, prior checking, and specification of priors in predictive spaces. We welcome applicants looking to work on any of these aspects and contribute to both theoretical development and practical software for aiding the prior specification process.

Keywords: Prior elicitation, Bayesian workflow, priors on predictive space, default priors

Supervision: Profs. Aki Vehtari (Aalto University) and Arto Klami (University of Helsinki); primary/secondary depending on the candidate’s interests

 More information

How Can I Apply?2022-11-17T09:12:03+02:00

In order to find the best match between candidates and supervisors, all applicants need to describe their research interests and how they see themselves contributing to the project which they are applying.

The applications are to be submitted through the eRecruitment system. By applying through our eRecruitment system, you can apply to multiple projects with one application.

Required attachments:

(1) Cover letter (1-2 pages in length) which clearly states your motivation for applying to work with said supervisor, how you could contribute to the research produced by the supervisor and research area, and other relevant information. For more senior researchers, you should also include a preliminary research plan (3-5 pages) which includes background information, research objectives and methods,  possible hypothesised results, possible collaboration, and any other key information.

(2) CV

(3) List of publications (please do not attach full copies of publications)

(4) A transcript of doctoral studies and the degree certificate of the doctoral degree

(5) Contact details of possible referees from 2-3 senior academic people. We will contact your referees, if recommendation letters are required.

All materials should be submitted in English in a PDF format. Note: files should be 5MB max. You can upload up to five files to the eRecruitment system, each 5MB.

In the evaluation of the candidates, emphasis is put on the quality of the candidate’s previous research and international experience as well as the innovativeness, impact potential, and feasibility of the research plan.

Candidates short-listed may be invited for an interview either at the Otaniemi or Kumpula campus or for an interview conducted via Zoom.


The candidate should have a PhD in an appropriate field at the time of starting the position. If a candidate does not have a PhD at the time of the application, a detailed plan of completion must be submitted. The candidate should also have an excellent track record in scientific research with publications in one or more fields relevant to the position. A good command of English is a necessary prerequisite.

Compensation, working hours, and place of work2022-11-17T09:32:13+02:00

HIIT follows the guidelines of the university agreement for salaries. The salary for a postdoctoral researcher starts typically from about 3500 EUR per month depending on experience and qualifications.  Those who are a more senior candidate shall receive a higher compensation in regards to their experience. The length of the contract as well as the start and end dates are negotiable. In addition to the salary, the contract includes occupational health benefits in combination with Finland’s comprehensive social security system. More information about the benefits from our hosting institutions can be found on their respective websites: Aalto University  University of Helsinki

The annual total workload of teaching/research staff at the recruiting universities is 1624 hours. The position is located at either Aalto University’s Otaniemi campus or the University of Helsinki’s Kumpula campus. The positions belong to the university career system and the selected persons will be appointed for fixed-term positions, for Postdoctoral Researchers typically for two years (unless otherwise stated) and for Research Fellows typically three years, with an option for renewal.  In addition to research work, the persons hired are expected to participate in the supervision of students and teaching following the standard practices of the hiring department.

Contact Information2022-11-17T09:36:20+02:00

For any question regarding the electronic application system, please contact Maaria Ilanko (

For questions regarding these positions, please contact supervising professor listed at or