Annual Report 2021: Highlights

In 2021, HIIT directly supported a wide range of collaborative research projects, but, due to the COVID-19 pandemic, fewer events and research visits were funded. Below we present a sample of activities of the Helsinki ICT community in 2021.

Enhancing nearest neighbor search with machine learning

Scaling up machine learning methods to deal with big data scenarios requires solving many challenging algorithmic problems. However, machine learning can also help speed up algorithms. A case where machine learning and algorithms research benefit each other is nearest neighbor search.

This work was partially funded by HIIT community support.

See the story.

Academy of Finland grants funding for FCAI’s second term

The Academy of Finland announced its funding for the second terms of its Finnish Flagship programs. The overall amount of funding reserved for the Finnish Center for Artificial Intelligence FCAI is up to over 11 million euros, of which a half has now been officially confirmed and the latter half will be granted early 2025.

See the story.

Harnessing psychological theory for better video games

The European Research Council (ERC) has funded Assistant Professor Elisa Mekler’s TheoryCraft project that aims to help both video game researchers and practitioners better understand and apply theories from psychology.

See the story.

NVIDIA AI Technology Center makes researchers better equipped to develop computationally demanding AI applications

Artificial intelligence researchers have benefited from NVIDIA’s expertise in utilizing graphics processing units (GPUs) and AI software in AI applications. NVIDIA has also contributed to research carried out at FCAI.

See the story.

Deep generative models for text-to-image generation

Deep generative models have achieved impressive results in the text-to-image generation task which is the task of generating a visual scene based on its textual description. The Master of Science thesis by Andrey Sukhobok studied the architectural choices of the transformer-based generative model DALL-E.

The work was partially funded with HIIT Community Support.

See the story.

Researchers teach artificial intelligence to be fluent in Finnish dialects

Finnish dialects create a lot of trouble when interacting with computers, since it is impossible to speak a language without speaking in a dialect of some sort. A research group has built artificial intelligence models that can automatically detect, normalize and generate Finnish dialects.

See the story.

ELLIS Unit Helsinki expands its base and grows its activities

ELLIS (the European Laboratory for Learning and Intelligent Systems) is a network of excellence that brings together top AI researchers, with the goal that the best basic research is done in Europe. A new twin unit in Manchester will open up opportunities for collaboration and exchange.

See the story.

Harnessing FCAI research for sustainability

The new program looks for ways that AI research can help us build a more sustainable future, and helps research projects communicate their contribution.

See the story.

New HIIT focus area: Foundations of Computing

Foundations of Computing is a new HIIT focus area pursuing research into fundamental methods and the mathematics of computation.

See the story.

Helsinki Brain & Mind utilizes digitalization and develops AI-based brain imaging analysis

In the coming years, the network in neuroscience research will focus its efforts on the development of digital solutions and innovations. The network’s new project, with a budget of over €880,000, has received substantial EU funding for the term 2021–2023.

See the story.

Artificial intelligence speeds up the creation of fusion energy

Fusion energy is no longer a project for the distant future. The current official EU aim is 35 years, and with AI methods, fusion is here even faster than would otherwise be possible.

See the story.

AI Finland & AI Day 2021: Reboot with AI

AI Day is the biggest annual event highlighting the frontrunner AI research in Finland. The event brings together researchers, companies, students and the public sector involved in the fast-developing field of AI.

See the story.

Novel brain model could help elucidate causes of neurodegenerative disorders

Researchers at Aalto have developed novel brain models to simulate functional networks corresponding closely to those observed in Human Neuroscience data. These models will help pinpoint disrupted brain mechanisms underlying neurodegenerative disorders, such as Alzheimer’s Disease.

See the story.

FCAI’s director Samuel Kaski receives major international funding for the development of AI methods

Samuel Kaski was awarded the Turing AI World-Leading Researcher Fellowship to establish a new center of excellence.

See the story.

Using visualization to support insight generation and sharing

Researchers from the Ubiquitous Interaction research group at the University of Helsinki devised a Carbon Dioxide Explorer that enables users to refer to visualization components (entities) in externalizing their insights. The results were used to apply for SRC funding for a project on data literacy.

This work was partially funded by HIIT Community Support.

See the story.

EntityBot is a digital assistant that knows users’ actual needs

The digital assistant EntityBot could help make computer usage easier by making suggestions based on the user’s actual needs, such as recommending relevant information.

See the story.

Transforming dextrous robotic grasping

FCAI researchers have recently developed two fast AI methods for grasping objects with multi-finger robotic hands, bringing us closer to the practical use of multi-finger robotic hands.

See the story.

How is Instagram used by the leading health agencies to engage with the public during the COVID-19 crisis?

A study by researchers at Aalto showed that most of the images shared by major health agencies relate to preventive measures, health advisories, gratitude and resilience of front-line workers. Social media messaging depicting well known personalities, influencers, and celebrities received greater engagement. Furthermore, posts with inquisitive messaging, infographics, or dispelling myths/fake news/misinformation also receive more attention.

See the story.

Computers can predict our preferences directly from our brain

A research team from Helsinki and Copenhagen demonstrates it is possible to predict individual preferences based on how a person’s brain responses match up to others. This could potentially be used to provide individually-tailored media content — and perhaps even to enlighten us about ourselves.

See the story.

Following e-cigarette conversations on Twitter using artificial intelligence

Machine learning methods help researchers follow social media promotions of addictive products to teenagers and young people.

See the story.

Artificial intelligence – a key to sustainable smart cities and mobility

Cities and transport are major contributors to CO2 emissions. FCAI’s recent webinar brought industry and academia together to discuss the need for new sustainable solutions, enabled by artificial intelligence.

See the story.

Janne Lindqvist was a guest on Aalto’s podcast ‘Ratkaisujen tarpeessa’

Janne Lindqvist and Miapetra Kumpula-Natri had a discussion about security in the digital world on Aalto’s podcast Ratkaisujen Tarpeessa (in Finnish).

See the story.

More certainty for deep learning

With the support of Saab, Aalto doctoral student Lassi Meronen is developing models for deep learning that know how to estimate their own uncertainty computationally. These models can be used, for example, in radars and medical diagnosis.

See the story.

AI learns to type on a phone like humans

Researchers at Aalto created a new computational model precisely replicates the eye and finger movements of touchscreen users – could lead to better auto-correct and keyboard usability for unique needs.

See the story.