Studia Generalia: Mikä on Big Data?

Lecturer : 
Salla-Maaria Laaksonen
Event type: 
Event
Doctoral dissertation
Respondent: 
Opponent: 
Custos: 
Event time: 
2016-10-06 17:00 to 19:00
Place: 
Yliopistonkatu 3
Description: 

Salla-Maaria Laaksonen puhuu Helsingin yliopiston Studia Generalia-luentosarjassa aiheella Digivaalit 2015: Mitä isot digitaaliset aineistot kertovat yhteiskunnasta ja vaikuttamisesta?

Digivaalit 2015 on Helsingin yliopiston Viestinnän tutkimuskeskus CRCn ja Tietotekniikan tutkimuslaitos HIITin yhteinen tutkimushanke.

Call for Participation: WITMSE 2016

Event type: 
Conference
Event time: 
2016-09-19 09:00 to 2016-09-21 16:00
Place: 
Siltavuorenpenger, Helsinki
Description: 
The Ninth Workshop on Information Theoretic Methods in Science and Engineering (WITMSE) will be held in Helsinki, Finland, on September 19–21, 2016. The workshop will cover hot topics at the intersection of statistics, information theory, machine learning, and their applications. The technical program will include plenary lectures and invited talks.
 
Plenary speakers
  • Wojciech Szpankowski, Purdue: Analytic Pattern Matching: From DNA to Twitter
  • John Shawe-Taylor, University College London: Learning to Detect an Immune Response from T-Cell Sequences
  • Andrew Barron, Yale: High-Dimensional Neural Networks: Statistical Information Theory and Computational Properties
The workshop is organized jointly by the Department of Computer Science of the University of Helsinki, the Helsinki Institute for Information Technology HIIT.
 
Please register at www.helsinki.fi/witmse2016. The registration charge is 185EUR (students 115EUR).

HIIT Kumpula Seminar: Secure genome sequence search based on homomorphic encryption

Lecturer : 
Prof. Kana Shimizu, Waseda University, Japan
Event type: 
HIIT seminar
Doctoral dissertation
Respondent: 
Opponent: 
Custos: 
Event time: 
2016-08-26 10:15 to 11:00
Place: 
Exactum B119
Description: 

Speaker: Prof. Kana Shimizu

Affiliation: Department of computer science and engineering, Faculty of science and engineering, Waseda University, Japan

Short bio: Dr. Kana Shimizu is an associate professor of Waseda university. After receiving Dr.Eng. in computer science from Waseda University in 2006, she joined the National Institute of Advanced Industrial Science and Technology (AIST). She also worked at the Memorial Sloan-Kettering Cancer Center in 2013-2015 as Visiting Investigator. In April 2016, she started her lab at Waseda University. Her research interest mainly centers on algorithms for biological sequence analyses. Her recent interest also includes privacy-preserving datamining for biological/biomedical data analyses. URL: http://iskana.github.io/web/index.html

Title: Secure genome sequence search based on homomorphic encryption.

Abstract: The state-of-the-art DNA sequencer generates 160 Giga bases per day, which is hundreds of thousands times as large amount of data as the technology of 15 years ago can generate. The huge cost down in DNA sequencing has encouraged large-scale personal genome sequencing, which eventually spotlighted privacy issues in genomics. In our work, we assumed the frequent case such that the user wish to query the server while hiding the contents of the query, and developed a novel algorithm that enables searching on DNA sequences without leaking user’s query to the server. The proposed algorithm combines a searchable string data structure such as (positional) Burrows-Wheeler Transform and a cryptographic technique called oblivious transfer, and allows variable length substring match. In an experiment using the dataset created from 1000 Genome project, our algorithm was order of magnitude efficient both in run time and data transfer overhead compared to the base line exhaustive method.

Guest Talk on The Economics of Cloud Computing

Lecturer : 
Liang Zheng
Event type: 
Event
Event time: 
2016-08-15 13:00 to 14:00
Place: 
Kumpula, Exactum, C 222
Description: 

The Economics of Cloud Computing
by Dr. Liang Zheng, Princeton University, Electrical Engineering Dept.
August 15, 2016 @Exactum C 222 
Time: 13:00 - 14:00

Abstract: As a form of could computing, Infrastructure-as-a-Service (IaaS) provides shared computingresources over the Internet, and has revolutionized the way that computing resources are utilized: they are virtualized in units of instances associated with remote virtual machines with specified amounts of CPU, memory, storage, and other attributes. Users can then lease these cloud computing capacities and pay for the execution of their jobs only as they are using the resources, eliminating setup and maintenance costs for the physical machines. With the growth of cloud services, cloud providers face highly dynamic user demands for their resources, making it difficult for them to maintain consistent quality-of-service (QoS).

We propose to use price incentives to stabilize user demands. Although cloud resources are often charged simply by usage-based pricing, it cannot handle in real-time the available capacity within datacenter networks and individual jobs' required instance hours and interruptibility. Thus, many cloud providers are turning to different pricing schemes to match their prices to real-time user demands. For example, auction-based pricing allows users to bid for spare cloud resources at a highly reduced rate, and volume-discount pricing encourages users' long-term usage by charging longer jobs a lower unit price. We provide insights into these pricing schemes by quantifying user demands with different prices, and derive optimal strategies to benefit both cloud providers and users.

Bio: Liang Zheng is a postdoc in Department of Electrical Engineering at Princeton University. She received her Ph.D. in Computer Science from the City University of Hong Kong in 2015. Her research interests are in user behavior analytics, network economicscloud computing, communication networks, nonlinear optimization and its applications. She was a finalist of the Microsoft Research Asia Fellowship in 2013. She was a recipient of the CityU Outstanding Academic Performance Award and first-class Student Research Excellence Award.

HIIT Kumpula Seminar: Energy-efficient communication and computation for IoT

Lecturer : 
Pulkit Grover
Event type: 
HIIT seminar
Event time: 
2016-08-19 10:15 to 11:00
Place: 
Exactum B119
Description: 

Title: Energy-efficient communication and computation for IoT:  fundamental limits, efficient strategies, and application to biosensing wearables

Abstract: How do we minimize energy required in short-distance communications? What is the minimum energy required to compute reliably using error and delay-prone gates or processors? With the advent of IoT and saturation of Moore's law (and Dennard's scaling), these questions have becoming increasingly important as researchers seek technologies for high-speed low-energy communications, and efficient alternatives to ultra-reliable CMOS devices. I’ll talk about our work on both fundamental limits on energy requirements (and how Shannon theory changes when computation is brought in), as well as novel strategies and architectures for minimizing communication and computation energy. This includes new coding techniques as well as strategies that perform reliable machine-learning on error/delay-prone and energy-limited components and sensors. Finally, I’ll talk about application of these ideas in design and implementation of IoT for noninvasive biopotential measurement, e.g. for neural interfaces. I'll discuss how a novel “hierarchical” architecture that limits error-accumulation turns out to have a substantially improved information-energy dissipation tradeoff than simply “compressing innovations” (a strategy known to be suboptimal from a work of Kim and Berger). This is a part of a larger work on utilizing information theory to motivate and engineer ultra-high-density neural sensing interfaces, as well as provide fundamental limits on their precision and performance.

Bio: Pulkit Grover (Ph.D. UC Berkeley'10, B.Tech.'03, M.Tech.'05 IIT Kanpur) is an assistant professor at CMU (2013-), working on information theory, circuit design, and biomedical engineering. His main contributions to science are towards developing a new theory of information (fundamental limits and practical designs) for low-energy communication, sensing, and computing by incorporating novel (noisy and noiseless) circuit-energy models to add to classical communication or sensing energy models. To apply these ideas to a variety of problems including communication, computing, sensing, and novel biomedical systems, his lab works extensively with circuit engineers, neuroscientists, and doctors. Pulkit is the recipient of the 2010 best student paper award at the IEEE Conference in Decision and Control (CDC); a 2010 best student paper finalist at the IEEE International Symposium on Information Theory (ISIT); the 2011 Eli Jury Dissertation Award from UC Berkeley; the 2012 Leonard G. Abraham best journal paper award from the IEEE Communications Society; a 2014 best paper award at the International Symposium on Integrated Circuits (ISIC); a 2014 NSF CAREER award; and a 2015 Google Research Award.

Pages