Annual digest of the research activities of the Faculty of Computer Science.
Conferences
NAACL
- Batch-Softmax Contrastive Loss for Pairwise Sentence Scoring Tasks (Anton Chernyavskiy, Dmitry Ilvovsky, Pavel Kalinin, Preslav Nakov)
- Towards Computationally Feasible Deep Active Learning (Akim Tsvigun, Artem Shelmanov, Gleb Kuzmin, Leonid Sanochkin, Daniil Larionov, Gleb Gusev, Manvel Avetisian, Leonid Zhukov)
ICML
- Secure Distributed Training at Scale (Eduard Gorbunov, Alexander Borzunov, Michael Diskin, Max Ryabinin)
- From Dirichlet to Rubin: Optimistic Exploration in RL without Bonuses (Daniil Tiapkin, Denis Belomestny, Eric Moulines, Alexey Naumov, Sergey Samsonov, Yunhao Tang, Michal Valko, Pierre Menard)
- The power of first-order smooth optimization for black-box non-smooth problems (Alexander Gasnikov, Anton Novitskii, Vasilii Novitskii, Farshed Abdukhakimov, Dmitry Kamzolov, Aleksandr Beznosikov, Martin Takac, Pavel Dvurechenskii, Bin Gu)
NeurIPS
- Towards Practical Computation of Singular Values of Convolutional Layers (Alexandra Senderovich, Ekaterina Bulatova, Anton Obukhov, Maxim Rakhuba)
- On Embeddings for Numerical Features in Tabular Deep Learning (Yury Gorishniy, Ivan Rubachev, Artem Babenko)
- SketchBoost: Fast Gradient Boosted Decision Tree for Multioutput Problems (Leonid Iosipoi, Anton Vakhrushev)
- Local-Global MCMC kernels: the best of both worlds (Sergey Samsonov, Evgeny Lagutin, Marylou Gabrié, Alain Durmus, Alexey Naumov, Eric Moulines)
- BR-SNIS: Bias Reduced Self-Normalized Importance Sampling (Gabriel Cardoso, Sergey Samsonov, Achille Thin, Eric Moulines, Jimmy Olsson)
- Optimistic Posterior Sampling for Reinforcement Learning with Few Samples and Tight Guarantees (Daniil Tiapkin, Denis Belomestny, Daniele Calandriello, Eric Moulines, Remi Munos, Alexey Naumov, Mark Rowland, Michal Valko, Pierre Ménard)
- Training Scale-Invariant Neural Networks on the Sphere Can Happen in Three Regimes (Maxim Kodryan, Ekaterina Lobacheva, Maksim Nakhodnov, Dmitry Vetrov)
- HyperDomainNet: Universal Domain Adaptation for Generative Adversarial Networks (Aibek Alanov, Vadim Titov, Dmitry Vetrov)
- Accelerated Primal-Dual Gradient Method for Smooth and Convex-Concave Saddle-Point Problems with Bilinear Coupling (Dmitry Kovalev, Alexander Gasnikov, Peter Richtarik)
- The First Optimal Acceleration of High-Order Methods in Smooth Convex Optimization (Dmitry Kovalev, Alexander Gasnikov)
- The First Optimal Algorithm for Smooth and Strongly-Convex-Strongly-Concave Minimax Optimization (Dmitry Kovalev, Alexander Gasnikov)
- Clipped Stochastic Methods for Variational Inequalities with Heavy-Tailed Noise (Eduard Gorbunov, Marina Danilova, David Dobre, Pavel Dvurechenskii, Alexander Gasnikov, Gauthier Gidel)
- A Damped Newton Method Achieves Global O(1/k^2) and Local Quadratic Convergence Rate (Slavomír Hanzely, Dmitry Kamzolov, Dmitry Pasechnyuk, Alexander Gasnikov, Peter Richtarik, Martin Takac)
- Distributed Methods with Compressed Communication for Solving Variational Inequalities, with Theoretical Guarantees (Aleksandr Beznosikov, Peter Richtarik, Michael Diskin, Max Ryabinin, Alexander Gasnikov)
- Optimal Algorithms for Decentralized Stochastic Variational Inequalities (Dmitry Kovalev, Aleksandr Beznosikov, Abdurakhmon Sadiev, Michael Persiianov, Peter Richtarik, Alexander Gasnikov)
- Optimal Gradient Sliding and its Application to Optimal Distributed Optimization Under Similarity (Dmitry Kovalev, Aleksandr Beznosikov, Ekaterina Borodich, Alexander Gasnikov, Gesualdo Scutari)
- Decentralized Local Stochastic Extra-Gradient for Variational Inequalities (Aleksandr Beznosikov, Pavel Dvurechenskii, Anastasiia Koloskova, Valentin Samokhin, Sebastian Stich, Alexander Gasnikov)
- Nonparametric Uncertainty Quantification for Single Deterministic Neural Network (Nikita Kotelevskii, Aleksandr Artemenkov, Kirill Fedyanin, Fedor Noskov, Alexander Fishkov, Artem Shelmanov, Artem Vazhentsev, Aleksandr Petiushko, Maxim Panov)
EMNLP
- PAUQ: Text-to-SQL in Russian (Daria Bakshandaeva, Oleg Dmitrievich Somov, Ekaterina Dmitrieva, Vera Davydova, Elena Tutubalina)
Д. Бакшандаева, О. Сомов, Е. Дмитриева, В. Давыдова, Елена Тутубалина)
- TAPE: Assessing Few-shot Russian Language Understanding (Ekaterina Taktasheva, Tatiana Shavrina, Alena Fenogenova, Denis Shevelev, Nadezhda Katricheva, Maria Tikhonova, Albina Akhmetgareeva, Oleg Zinkevich, Anastasiia Bashmakova, Svetlana Iordanskaia, Alena Spiridonova, Valentina Kurenshchikova, Ekaterina Artemova, Vladislav Mikhailov)
- Active Learning for Abstractive Text Summarization (Akim Tsvigun, Ivan Lysenko, Danila Sedashov, Ivan Lazichny, Eldar Damirov, Vladimir Karlov, Artemy Belousov, Leonid Sanochkin, Maxim Panov, Alexander Panchenko, Mikhail Burtsev, Artem Shelmanov)
Other papers by the staff and students of the Faculty can be found here.
On 16–20 May Professor Ivan Arzhantsev presented a course Cox Rings with Applications in Algebraic Geometry at the Invariants in Algebraic Geometry spring school in Dijon, France. At the same school Kirill Shakhmatov gave a talk Unipotent group actions on complete toric varieties.
On 19 July German Magaj presented his poster Geometric and topological properties of the internal representation of neural networks at the Young Topologists Meeting 2022 in Copenhagen, Denmark.
On 18–23 September the researchers of the Laboratory on Algebraic Transformation Groups presented at the conference Algebraic Groups, their Friends and Relations at the St. Petersburg Department of Steklov Mathematical Institute of Russian Academy of Sciences. Ivan Arzhantsev presented an invited talk Tits-Type Alternative Automorphism Groups, while Alexander Perepechko gave a talk Automorphism Groups of Rigid Affine Surfaces: the Identity Component.
Events
On 14–18 February, the first international Winter School on Software Engineering took place. A total of 837 participants from all over the world registered for the school. The school included lectures by representatives of the Schaffhausen Institute of Technology, IT University of Copenhagen, JetBrains, AliExpress, University of Liverpool, and other representatives of academia and the industry. Recorded lectures of the school are available.
On 25 February the annual CERN workshop on high energy physics for high school students was held. It was organised by LAMBDA researchers Fedor Ratnikov and Denis Derkach.
On 25–26 May the first part of ABZ Meeting 2022 online conference was held. Maria Poptsova and Alan Herbert, reseach adviser of the international bioinformatics laboratory, helped to organise the event along Molecules, RNA Society, International Journal of Molecular Sciences, and InsideOutBio. 2501 participants from 24 countries took part in the conference. The second part of the event is planned to take place in September.
On 1–3 Septemer an international conference Computer Methods of Cognitome Analysis took place. It was held by the International Laboratory of Algebraic Topology and Its Applications. Programme and recorded lectures of the conference are available.
On 1–3 November the first autumn conference and school on machine learning Fall into ML took place. It was organised by the International Laboratory of Stochastic Algorithms and High-Dimensional Inference and the Laboratory of Methods for Big Data Analysis with the support of HSE AI Centre and the Russian Science Foundation. Programme of the school and conference, and recorded lectures are available.
Regular seminars
Colloquium
- 1 February Approximation with neural networks of minimal size: exotic regimes and superexpressive activations (Dmitry Yarotsky)
- 1 March Power Indicies for Attribution of JSM-hypotheses and Formal Concepts (Dmitry Ignatov)
- 19 April Statistics in Tandem Mass Spectrometry Data Analysis (Attila Kertes-Farkas)
- 24 May Do process models behave identically? Algorithmics and Decidability of Bisimulation Equivalences (Irina Lomazova)
- 6 December Polynomial Chaos Expansion and Machine Learning: Benefits, Challenges, Applications (Alexandr Tarakanov)
International Laboratory for Intelligent Systems and Structural Analysis
- 15 April "Less is more" based heuristics for Minimum sum-of-squares clustering" (Nenad Mladenovich, UAE)
- 11 May Informative discourse feature selection for analysis of textual data (Elizaveta Goncharova)
- 27 May HAO Intelligence for Big Wisdom (Xindong Wu, Hefei University of Technology, China)
Laboratory of Theoretical Computer Science
- 10 February The number of homomorphisms into finite algebras (Libor Barto, Charles University, Czech Republic)
- 17 February On the Generative Power of Semi-Bracketed Contextual Grammars (Anand Mahendran)
- 17 March Fitting Protein into H-Core: Structural Analysis of Discrete Hydrophobic Cores (Andrey Ignatov)
- 24 April Evolution of a bipartite random graph (Pavel Zakharov)
Structural Learning Seminar
- 19 September Statistics of McKean-Vlasov equations (Senis Belomestny)
- 26 September Multi-Level Unbiased MCMC Estimators for Discretized Models (Alexander Tarakanov)
- 3 October Is Q-learning Provably Efficient? (Article overview) (Yana Hassan)
- 24 October The Primacy Bias in Deep Reinforcement Learning (Evgenii Nikishin, Mila, Université de Montréal)
- 31 October A Contrastive Approach to Online Change Point Detection (Valeria Sherbakova)
- 7 November Theoretical guarantees for approximate sampling from smooth and log-concave densities (Artur Goldman)
- 14 November Rapid Convergence of the Unadjusted Langevin Algorithm: Isoperimetry Suffices (Daria Demidova)
- 21 и 28 November Diffusion Probabilistic Models explosion (Anton Zubekhin)
- 5 December Log-concave sampling: Metropolis-Hastings algorithms are fast (Varvara Rudenko)
- 12 December Exponential Concentration Inequalities for Additive Functionals of Markov Chains (Marina SheShukova)
- 19 December Sharper Bounds for Uniformly Stable Algorithms (Timofey Grinev)
International Laboratory of Algebraic Topology and Its Applications
- 17 February Wasserstein Stability for Persistence (Primoz Skraba, Queen Mary University of London)
- 18 March Path homology and join of digraphs (A. A.Grigor'yan, Bielefeld University, Germany)
- 25 March Simplicial complexes of polyomino tilings (Djordje Baralić, Mathematical Institute of the Serbian Academy of Sciences and Arts)
- 8 April Configuration space of Moebius Kaleidocycle (Shizuo Kaji, Kyushu University, Japan)
- 15 April Parameterized Vietoris-Rips Filtrations via Covers (Bradley Nelson, University of Chicago)
- 20 May Mathematical AI for molecular data analysis Kelin Xia, Nanyang Technological University, Singapore)
- 15 July Spectral clustering of combinatorial fullerene isomers based on their facet graph structure (Artur Bille, Ulm University)
- 21 October A quadratic estimation for the Kühnel conjecture (Svyatoslav Genger, MIPTи Arkady Skopenkov, MIPT)
- 18 November Cup products of the cohomology of 4-dimensional toric orbifolds (Tse Leung So, Western University, Canada)
Laboratory of Process-Aware Information Systems
- 21 January Data Petri Nets as a Formalism for Representing Decision-Aware Process Models (Nikolay Suvorov)
- 8 September Studying the properties of timed finite state machines (Evgeny Vinarskii)
- 15 September Overview of real-time models (Evgeny Vinarskii)
Laboratory of Complex Systems Modelling and Control
- 26 May Birth of Strange nonchaotic Attractors Nonlinear Dynamical Systems (Suresh Kumarasamy)
- 2 June Neural networks and its application (R. Vadive)
- 9 June Networked Control Systems and Its Applications (Nallappan Gunasekaran)
- 13 October Management of Localized Waves in Inhomogeneous Media (K. Sakkaravarthi)
- 27 October Impacts of Global Climate Change - 2022: Eagle's Eye View (R. Mahendran)
- 10 November Magnetization Switching in Pentalayer Nanopillar Alloys with Oscillatory Interlayer Exchange Coupling (Dr. D. Aravinthan)
Laboratory on AI for Computational Biology
- 1 September LC-MS-based lipidomics: data analysis challenges and applications to psychiatric disorder research (Anna Tkachev, Skoltech)
- 12 October QA on Data Independent Acquisition for tandem mass spectrometry (Vadim Demichev, Charite University Medicine Berlin)
- 13 October Deep learning applications in proteomics mass spectrometry (Prof. William Stafford Noble, University of Washington)
- 17 October Machine learning based approaches for analysis of cancer genome breakpoints (Ksenia Tcheloshina)
- 19 October DIAmeter: matching peptides to data-independent acquisition mass spectrometry data (Yang Lu, University of Washington)
- 28 November Light scattering techniques – versatile tools to study nanoparticle dispersions (István Szilágyi, University of Szeged)
Bayesian Methods of Machine Learning Seminar
- 25 February Formalism of quantum mechanics from the point of view of machine learning (Alexander Lobashev)
- 11 March A Random Matrix Theory approach to Deep Learning optimisation and generalisation (Diego Granziol, Huawei London AI Theory)
- 18 March Denoising Diffusion Restoration Models (Tingir Badmayev, MSU)
- 25 March Mathematics of Multi-Antenna Transmission in 5G networks (Ilya Sinilshchikov, Evgeny Bobrov)
- 8 April Autoformer and Autoregressive Denoising Diffusion Models for Time Series Forecasting (Nikita Gushchin)
- 15 April Bilateral Denoising Diffusion Models for Fast and High-Quality Speech Synthesis (Ivan Shchekotov, Samsung AI Center)
- 22 April MLP архитектура для решения задач CV и NLP (Dmitry Kurtsev, MSU)
- 20 May Neural Optimal Transport (Alexander Korotin, Skoltech, AIRI)
- 16 September Discrete Disentangled Representations for Object-Centric Visual Tasks (Alexander Panov, Daniil Kirilenko, Alexey Kovalev, AIRI)
- 14 October Overview of control techniques for text and image models (Sergey Troshin)
- 21 October Training Scale-Invariant Neural Networks on the Sphere Can Happen in Three Regimes (Maxim Kodryan)
- 18 November Grokking: Generalization Beyond Overfitting on Small Algorithmic Datasets (Timofey Uzhakov)
- 25 November Discovering faster matrix multiplication algorithms with reinforcement learning (Alexander Novikov, DeepMind)
- 9 December Domain Adaptation of GANs (Aibek Alanov)
Algebraic Transformation Groups Seminar
- 19 January Polytopal linear groups (Anton Shafarevich)
- 2 February The commuting derivations conjecture (Veronika Kikteva)
- 4 February Existence results for B-root subgroups on affine spherical varieties (Roman Avdeev)
- 16 February Characterization of affine toric varieties by their automorphism groups (Andriy Regeta)
- 2, 16 March Isotropy group of an LND of the polynomial algebra in three variables (Sergey Gaifullin)
- 6 April Automorphism groups of affine varieties without non-algebraic elements (Alexander Perepechko)
- 20 April Hessenberg varieties (Ekaterina Presnova)
- 11 May Wild automorphisms of graded algebras (Anton Trushin)
- 25 May Map enumeration problem (Veronika Treumova)
- 15 June Isometries of Mukai lattices (Ivan Bel'diev)
- 9 November Matroid Schubert varieties as equivariant compactifications of affine spaces (Colin Crowley, University of Wisconsin–Madison)
- 14 December Characterization of algebraic varieties by their groups of symmetries (Alvaro Liendo, Universidad de Talca)
Mirror Laboratory (LAMBDA)
- 1 June In silico discovery of novel topological materials (Oleg Yaziev, EPFL)
- 10 June Generative flow networks for modeling highly structured objects (Anastasia Volokhova, Mila & Université de Montréal)
Open seminars
- 24 February Non-Smooth Optimization with Noisy Gradient-Free Oracle (Darina Dvinskikh, Kharkevich Institute for Information Transmission Problems of the Russian Academy of Sciences)
- 24 February Utilization of Polynomial Chaos Expansion for acceleration of Optimal Bayesian Experimental Design (Alexander Tarakanov, Huawei R&D Moscow)
New departments and projects
In March, a joint department with Sber opened at the Faculty. It is headed by Alexay Masutin, who is also the head of the HSE AI Center.
In June, the Faculty of Computer Science has opened a joint department with Tinkoff. The department courses will enrol about 80 students annually, who will get internships at Tinkoff, with the best-achieving students invited for interview at Tinkoff. The department is headed by Andrey Ivanov.Polyhedral products in topology, geometry and combinatorics (Ivan Limonchenko)
Doctors of Science
- Attila Kertesz-Farkas
- Alexey Naumov
- Alexey Ossadchii
- Elena Gryazina
.png "Research Digest 2022")