Categories
Allgemein

BIFOLD Colloquium 04/07/2022

Home >

BIFOLD Colloquium 04/07/2022

Algorithms for inferring cancer evolution from haplotype-specific somatic copy-number alterations

Speaker: Prof. Dr. Roland F. Schwarz, Center for Integrative Oncology (CIO), Cancer Research Center Cologne Essen (CCCE), University Hospital and University of Cologne
Date & Time: July 4, 2022; 4:00 pm
Venue: TU Berlin, Straße des 17. Juni 135, 10623 Berlin, Main building, Room: H 1028

Abstract: Traditionally, phylogenetic inference methods have mostly focused on inferring evolutionary trees from single nucleotide variants (SNVs). In cancer, in addition to SNVs, genomic rearrangements and somatic copy number alterations (SCNAs) play an important role in the development and progression of the disease, and SCNAs can provide a rich source of genetic variation suitable for reconstructing cancer evolution. SCNAs thereby pose specific algorithmic challenges owing to the non-independence of adjacent genomic loci and their overlapping and cascading nature. Additionally, accurate phylogenetic inference from SCNAs requires identifying the parental chromosome of origin of each evolutionary event, a non-trivial task in many short-read sequencing datasets.
Roland Schwarz will introduce two key algorithms for inferring cancer evolution from SCNA profiles derived from multi-region sequencing data: refphase, which identifies the parental chromosome of origin of SCNAs by leveraging heterozygous germline variants shared between multiple samples from the same patient, and MEDICC2, a complete phylogenetic inference algorithm for SCNA profiles. He will describe how his group overcomes the problem of phasing of SCNAs, leverage finite-state transducers to compute exact minimum event distances between pairs of SCNA profiles, and detect, order and place individual evolutionary SCNA events including whole-genome doublings on the phylogenetic tree. Finally, he is going to demonstrate how in the future, the event histories inferred by MEDICC2 might be leveraged by machine learning algorithms to derive copy-number signatures which can identify the mutational processes underlying individual cancer types.

Speaker: Dr. Roland F. Schwarz is Professor for Computational Cancer Biology at the Center for Integrative Oncology (CIO), Cancer Research Center Cologne Essen (CCCE), University of Cologne, Germany. He is a renowned expert in machine learning, theoretical computer science and clinical oncology. He uses machine learning and statistical algorithms to explore the causes and functional consequences of differences in tumors and cancer evolution. 2016 Schwarz was awarded the Prize of the Berlin-Brandenburg Academy of Sciences for excellence in cancer research for his work on cancer genome evolution.

Prof. Dr. Roland Schwarz
C: private

Categories
Allgemein

BIFOLD Colloquium 2022/06/13

Home >

BIFOLD Colloquium 2022/06/13

Perspectives on Data Stream Processing for Resource-constrained Devices


Speaker: Gabriele Mencagli, University of Pisa
Venue: Virtual
Date & Time: June, 13, 2022; 4:00pm


Registration: If you are interested in participating, please contact: coordination@bifold.berlin

Abstract:
Resource-constrained devices are widely diffused in highly-distributed computing environments like IoT platforms, edge and far-edge computing infrastructures, and fog scenarios. Such devices are usually equipped with low-power CPUs, and they often include integrated co-processors (e.g., GPUs and FPGAs) available as System-on-Chip architectures. Data Stream Processing is a hot research topic focusing on efficient data analysis techniques and processing systems to extract analytics, knowledge and, in general, perform generic computations on unbounded sequences of data flows arriving at high speed. Tradition streaming systems (e.g., Apache Storm, Flink, Spark Streaming) target scale-out scenarios (i.e., clusters of homogeneous high-end servers and Clouds).

The transition to efficiently support resource-constrained devices advocates to profitably configure the processing model of the existing streaming systems to fit at best the new hardware or, alternatively, to design new frameworks from scratch. This talk will show the research directions currently followed by the Parallel Programming Models (PPMs) group at the University of Pisa, Department of Computer Science.

Speaker:
Gabriele Mencagli is Associate Professor at the Department of Computer Science, University of Pisa, Italy. He is member of the Parallel Programming Models group, doing research in High Performance Computing, Parallel Programming and Data Stream Processing.
The main contribution of the group is the development of novel parallel programming frameworks for multicores and heterogeneous systems, with special focus on high-level abstractions (parallel patterns) for easing the effort to develop efficient parallel software on different kinds of hardware resources.

Categories
Allgemein

Art of Entanglement

Home >

Art of Entanglement

The Art of Entanglement

The Berlin Institute for the Foundations of Learning and Data (BIFOLD), together with the Science Gallery at Technische Universität Berlin, has announced a new artist in residence program called “Art of Entanglement”. The goal of the program is to combine artistic and scientific perspectives of artificial intelligence. The program is endowed with a gross total of 30,000 euros. The open call was published on sciencegallery.submittable.com. Applications are open to artists based in Berlin who are interested in working intensively with topics and scientists in the fields of Big Data Management and Machine Learning as well as their intersection. The selected artist will have the opportunity to realize an artistic project of their choice at BIFOLD, the national Berlin Center of Excellence for Artificial Intelligence at TU Berlin, and the Science Gallery platform.


The Berlin Competence Center BIFOLD was created in 2019 by merging and expanding the Berlin Big Data Center (BBDC) and Berlin Center for Machine Learning (BZML). BIFOLD conducts agile foundational research at the intersection of Big Data Management (BM) and Machine Learning (ML), the key innovation drivers of artificial intelligence (AI) applications.

The advertised artist in residence program establishes the partnership between BIFOLD and the Science Gallery initiative in Berlin. TU Berlin is the newest member of the Science Gallery International network – the world’s only university network dedicated to public engagement with science and art. The selection process and the residency are curated and accompanied by the Austrian curator Dr. Claudia Schnugg. Claudia Schnugg is an experienced curator in the field of art and science. Parallel to her curatorial work, she conducts research at the intersections of art and aesthetics with science, technology and organizations. “The scientific knowledge and potential of these new technologies is constantly growing. In parallel, BIFOLD is keen to fuel social, cultural and artistic engagement with these topics,” explains BIFOLD Director Prof. Dr. Klaus-Robert Müller.

A new artist in residence program explores the interplay between AI research and art.
(Copyright: BIFOLD)

“From the ‘Art of Entanglement’ program, we hope to gain new perspectives on how AI systems are used to solve problems, but also how they can undermine or highlight asymmetries and inequalities,” adds Prof. Dr. Volker Markl, also BIFOLD Director.

The residency program provides a framework to explore the role of the artist, art and culture in the rapidly evolving wave of AI innovation in the vibrant AI metropolis of Berlin. The open call closes on April 25, 2022. By the end of May, a jury consisting of curator Dr. Claudia Schnugg, members of BIFOLD, Science Gallery Berlin and external experts will select and notify the artist. The actual residency will begin in September 2022, and the artworks created during the residency will be exhibited starting in February 2023. The selected artist will receive a gross total of EUR 30,000 to cover the artist’s fee, time at BIFOLD, fees for presentations and workshops during the residency period, as well as material and production costs for the artwork.

More Information:

The open call is published here.

Categories
Allgemein

2020 Pattern Recognition Best Paper Award

Home >

2020 Pattern Recognition Best Paper Award

2020 Pattern Recognition Best Paper Award

A team of scientists from TU Berlin, Fraunhofer Heinrich Hertz Institute (HHI) and University of Oslo has jointly received the 2020 “Pattern Recognition Best Paper Award” and “Pattern Recognition Medal” of the international scientific journal Pattern Recognition. The award committee honored the publication “Explaining Nonlinear Classification Decisions with Deep Taylor Decomposition” by Dr. Grégoire Montavon and Prof. Dr. Klaus-Robert Müller from TU Berlin, Prof. Dr. Alexander Binder from University of Oslo, as well as Dr. Wojciech Samek and Dr. Sebastian Lapuschkin from HHI.

Dr. Grégoire Montavon with the 2020 Pattern Recognition Best Paper Award in hand.

The publication addresses the so-called black box problem. Machine Learning methods, in particular Deep Learning, successfully solve a variety of tasks. However, in most cases they fail to provide the information that has led to a particular decision. The paper tackles this problem by using a pixel-by-pixel decomposition of nonlinear classifications and evaluates the procedure in different scenarios. This method provides a theoretical framework for Explainable Artificial Intelligence (XAI) that is generally applicable. XAI is a major research field of the Berlin Institute for the Foundations of Learning and Data (BIFOLD), of which the authors from TU Berlin and HHI are members.

The award was presented to Grégoire Montavon in January 2021, during the virtual International Conference on Pattern Recognition (ICPR). The “Pattern Recognition Best Paper Award” is granted every two years. It recognizes a highly cited paper in the area of pattern recognition and its application areas such as image processing, computer vision and biometrics.

“We are very proud to receive this award and for our work to be highlighted within the global scientific community.”

Dr. Grégoire Montavon.

Categories
Allgemein

“Data is the new soil!” – Interview with Prof. Markl

Home >

“Data is the new soil!” – Interview with Prof. Markl

“Data is not the new oil, but the new soil!” – Newspaper Interview with Prof. Markl

In an interview with the German newspaper ‘Der Tagesspiegel,’ one of BIFOLD’s directors, Prof. Dr. Markl, explains the necessary steps to drive Europe forward in terms of data sovereignty and innovation ecosystems.

In an interview with Oliver Voss (Tagesspiegel), Prof. Dr. Volker Markl, Head of the DIMA (TUB) and IAM (DFKI) research groups and one of the directors of the BIFOLD institute, explains the challenges and opportunities in handling data as a production factor in Europe. How can we achieve AI leadership across Europe and how can we make progress in research, training and innovation in the areas of big data management and machine learning?

Data is not the new oil, but the new soil. Because just as new grain is created from the soil, new information can be gained from data. And just as I have to fertilize or water the soil, I also have to maintain, clean and update data. But above all, you can import raw materials, but you have to have a production factor in your own country, which is why sovereignty in this area is so important.

Prof. Dr. Volker Markl

Among many other points, Prof. Markl suggests to concentrate on excellence in research and education in European AI Centers such as BIFOLD and the other Institutions in the German Network of National Centers of Excellence for AI Research. The combination of strong research and education as well as an extensive data analysis infrastructure to share and process data across different domains may create the innovation ecosystem Europe needs to compete internationally.

Read the full interview in Tagesspiegel (in german).