02.04.2020, 11:00 Uhr, Math Machine Learning seminar MPI MIS + UCLA

• Yu Guang Wang (University of New South Wales, Sydney)
• Haar Graph Pooling
• see PDF Abstract
• see video

09.04.2020, 11:00 Uhr, Deep Learning Theory Group Seminar

• Pradeep Banerjee (MPI MIS, Leipzig)
• Non-vacuous PAC-Bayes generalization bounds

09.04.2020, 17:00 Uhr, Math Machine Learning seminar MPI MIS + UCLA

• Michael Arbel (University College London)
• Kernelized Wasserstein Natural Gradient
• see video

16.04.2020, 11:00 Uhr, Deep Learning Theory Group Seminar

• Pradeep Banerjee (MPI MIS, Leipzig)
• Information theoretic generalisation bounds

16.04.2020, 17:00 Uhr, Math Machine Learning seminar MPI MIS + UCLA

• Anton Mallasto (University of Copenhagen)
• Estimation of Optimal Transport in Generative Models
• see video

23.04.2020, 11:00 Uhr, Deep Learning Theory Group Seminar

• Hanna Tseran (MPI MIS, Leipzig)
• Neural Network Expressivity (Overview)

23.04.2020, 17:00 Uhr, Math Machine Learning seminar MPI MIS + UCLA

• Johannes Mueller (MPI MIS, Leipzig)
• Deep Ritz Revisited
• see video

30.04.2020, 11:00 Uhr, Deep Learning Theory Group Seminar

• Pierre Brechet (MPI MIS, Leipzig)
• Optimal transport with global invariance (Paper review)

30.04.2020, 17:00 Uhr, Math Machine Learning seminar MPI MIS + UCLA

• Quynh Nguyen (University Saarbruecken)
• Loss surface of deep and wide neural networks
• see video

07.05.2020, 11:00 Uhr, Deep Learning Theory Group Seminar

• Johannes Müller (MPI MIS, Leipzig)
• Space-time expressivity of ResNets

07.05.2020, 17:00 Uhr, Math Machine Learning seminar MPI MIS + UCLA

• Kathlén Kohn (KTH Royal Institute of Technology)
• The geometry of neural networks
• see video

08.05.2020, 17:00 Uhr, Math Machine Learning Group Seminar

• Yoni Dukler (University of California, Los Angeles)
• Deep Learning for Self Driving

14.05.2020, 11:00 Uhr, Deep Learning Theory Group Seminar

• Nina Otter (University of California, Los Angeles)
• Topology of deep neural networks

14.05.2020, 17:00 Uhr, Math Machine Learning seminar MPI MIS + UCLA

• Benjamin Fehrman (University of Oxford)
• Convergence rates for the stochastic gradient descent method for non-convex objective functions
• see video

15.05.2020, 17:00 Uhr, Math Machine Learning Group Seminar

• Thomas Merkh (University of California, Los Angeles)
• Recurrences and transfer methods for counting activation patterns in CCNs

21.05.2020, 11:00 Uhr, Deep Learning Theory Group Seminar

• Hanna Tseran (MPI MIS, Leipzig)
• Algebraic geometry and statistical learning theory (Overview)

21.05.2020, 17:00 Uhr, Math Machine Learning seminar MPI MIS + UCLA

• Dennis Elbrächter (Universität Wien)
• How degenerate is the parametrization of (ReLU) neural networks?
• see video

22.05.2020, 17:00 Uhr, Math Machine Learning Group Seminar

• Hui Jin (University of California, Los Angeles)
• Implicit bias of gradient descent on wide neural networks

28.05.2020, 11:00 Uhr, Deep Learning Theory Group Seminar

• Pierre Brechet (MPI MIS, Leipzig)
• KALE: When energy-based learning meets adversarial training (Paper review)

28.05.2020, 17:00 Uhr, Math Machine Learning seminar MPI MIS + UCLA

• Jonathan Frankle (Massachusetts Institute of Technology)
• The Lottery Ticket Hypothesis: On Sparse, Trainable Neural Networks

04.06.2020, 11:00 Uhr, Deep Learning Theory Group Seminar

• Pradeep Banerjee (MPI MIS, Leipzig)
• PAC Bayes and wide minima

04.06.2020, 17:00 Uhr, Math Machine Learning seminar MPI MIS + UCLA

• Ulrich Terstiege (Rheinisch-Westfälische Technische Hochschule Aachen)
• Learning deep linear neural networks: Riemannian gradient flows and convergence to global minimizers

11.06.2020, 11:00 Uhr, Deep Learning Theory Group Seminar

• Johannes Müller (MPI MIS, Leipzig)
• A closer look at the approximation capabilities of neural networks (Paper review)

11.06.2020, 17:00 Uhr, Math Machine Learning seminar MPI MIS + UCLA

• Poorya Mianjy (Johns Hopkins University)
• Understanding the Algorithmic Regularization due to Dropout
• see video

12.06.2020, 17:00 Uhr, Math Machine Learning Group Seminar

• Yonatan Dukler (University of California, Los Angeles)
• Optimization Theory for ReLU Neural NetworksTrained with Normalization Layers

18.06.2020, 11:00 Uhr, Deep Learning Theory Group Seminar

• Hanna Tseran (MPI MIS, Leipzig)
• Complexity of Linear Regions in Deep Networks - II (Paper Review)

18.06.2020, 17:00 Uhr, Math Machine Learning seminar MPI MIS + UCLA

• Alessandro Achille (University of California, Los Angeles)
• Structure of Learning Tasks and the Information in the Weights of a Deep Network
• see video (MPI MiS internal only)

25.06.2020, 11:00 Uhr, Deep Learning Theory Group Seminar

• Pierre Bréchet (MPI MIS, Leipzig)
• Probability functional descent: A unifying perspective on GANs, Variational inference, and Reinforcement learning (Paper review)

25.06.2020, 17:00 Uhr, Math Machine Learning seminar MPI MIS + UCLA

• Adam Gaier (INRIA)
• Weight Agnostic Neural Networks

02.07.2020, 17:00 Uhr, Math Machine Learning seminar MPI MIS + UCLA

• Wenda Zhou (Columbia University)
• New perspectives on cross-validation
• see video

03.07.2020, 17:00 Uhr, Math Machine Learning seminar MPI MIS + UCLA

• Kai Fong Ernest Chong (Singapore University of Technology and Design)
• The approximation capabilities of neural networks
• see video

10.07.2020, 17:00 Uhr, Math Machine Learning seminar MPI MIS + UCLA

• Nasim Rahaman (MPI-IS Tübingen, and Mila, Montréal)
• On the Spectral Bias of Neural Networks
• see video

20.07.2020, 17:00 Uhr, Mathematics of Data Seminar

• Franca Hoffmann (California Institute of Technology)
• Kalman-Wasserstein Gradient Flows

23.07.2020, 17:00 Uhr, Math Machine Learning seminar MPI MIS + UCLA

• Léonard Blier (Facebook AI Research, Université Paris Saclay, Inria)
• The Description Length of Deep Learning Models

30.07.2020, 11:00 Uhr, Deep Learning Theory Group Seminar

• Hanna Tseran (MPI MIS, Leipzig)
• Tropical Geometry for Counting Linear Regions (Progress Report)

30.07.2020, 17:00 Uhr, Math Machine Learning seminar MPI MIS + UCLA

• Robert Peharz (TU Eindhoven)
• Minimal Random Code Learning: Getting Bits Back from Compressed Model Parameters
• see video

31.07.2020, 11:00 Uhr, Deep Learning Theory Group Seminar

• Quynh Nguyen (MPI MIS, Leipzig)
• On connected sublevel sets in deep learning

31.07.2020, 17:00 Uhr, Math Machine Learning seminar MPI MIS + UCLA

• Greg Ongie (University of Chicago)
• A function space view of overparameterized neural networks
• see video

04.08.2020, 11:00 Uhr, Mathematics of Data Seminar

• Marco Mondelli (IST Austria)
• Understanding Gradient Descent for Over-parameterized Deep Neural Networks

06.08.2020, 11:00 Uhr, Deep Learning Theory Group Seminar

• Yuguang Wang (MPI MIS, Leipzig)
• Distributed Filtered Hyperinterpolation on Manifolds

06.08.2020, 17:00 Uhr, Math Machine Learning seminar MPI MIS + UCLA

• Simon S. Du (University of Washington)
• Ultra-wide Neural Network and Neural Tangent Kernel
• see video (MPI MiS internal only)

13.08.2020, 11:00 Uhr, Deep Learning Theory Group Seminar

• Johannes Müller (MPI MIS, Leipzig)
• The tangent kernel of parametric models: A unified approach (tentative)

13.08.2020, 17:00 Uhr, Math Machine Learning seminar MPI MIS + UCLA

• Lénaïc Chizat (CNRS - Laboratoire de Mathématiques d'Orsay)
• Analysis of Gradient Descent on Wide Two-Layer ReLU Neural Networks
• see video

14.08.2020, 17:00 Uhr, Math Machine Learning seminar MPI MIS + UCLA

• Ido Nachum (École Polytechnique Fédérale de Lausanne)
• Regularization by Misclassification in ReLU Neural Networks
• see video

20.08.2020, 11:00 Uhr, Deep Learning Theory Group Seminar

• Quynh Nguyen (MPI MIS, Leipzig)
• Some thoughts on proving convergence of gradient descent for neural networks

20.08.2020, 17:00 Uhr, Math Machine Learning seminar MPI MIS + UCLA

• Arthur Jacot (École Polytechnique Fédérale de Lausanne)
• Neural Tangent Kernel: Convergence and Generalization of DNNs
• see video

21.08.2020, 17:00 Uhr, Math Machine Learning seminar MPI MIS + UCLA

• Preetum Nakkiran (Harvard University)
• Distributional Generalization: A New Kind of Generalization
• see video

27.08.2020, 11:00 Uhr, Deep Learning Theory Group Seminar

• Yuguang Wang (MPI MIS, Leipzig)
• Decimated Framelet System and Fast \(\gph\)-Framelet Transforms on Graph

27.08.2020, 17:00 Uhr, Math Machine Learning seminar MPI MIS + UCLA

• Felix Draxler (Heidelberg University)
• Characterizing The Role of A Single Coupling Layer in Affine Normalizing Flows
• see video

03.09.2020, 11:00 Uhr, Deep Learning Theory Group Seminar

• Pierre Bréchet (MPI MIS, Leipzig)
• Wasserstein gradient flow

03.09.2020, 17:00 Uhr, Math Machine Learning seminar MPI MIS + UCLA

• Niladri S. Chatterji (UC Berkeley)
• Upper and lower bounds for gradient based sampling methods
• see video

04.09.2020, 11:00 Uhr, Deep Learning Theory Group Seminar

• Ido Nachum (École Polytechnique Fédérale de Lausanne)
• On the Information Complexity of Learning

04.09.2020, 17:00 Uhr, Math Machine Learning seminar MPI MIS + UCLA

• Nadav Cohen (Tel Aviv University)
• Analyzing Optimization and Generalization in Deep Learning via Dynamics of Gradient Descent
• see video

17.09.2020, 11:00 Uhr, Deep Learning Theory Group Seminar

• Johannes Müller (MPI MIS, Leipzig)
• Complexity reduction for optimal policies: the linear case

17.09.2020, 17:00 Uhr, Math Machine Learning seminar MPI MIS + UCLA

• Mahito Sugiyama (National Institute of Informatics, JST, PRESTO)
• Learning with Dually Flat Structure and Incidence Algebra

18.09.2020, 17:00 Uhr, Math Machine Learning seminar MPI MIS + UCLA

• Mert Pilanci (Stanford University)
• Exact Polynomial-time Convex Formulations for Training Multilayer Neural Networks: The Hidden Convex Optimization Landscape

24.09.2020, 17:00 Uhr, Math Machine Learning seminar MPI MIS + UCLA

• Liwen Zhang (University of Chicago)
• Tropical Geometry of Deep Neural Networks
• see video

25.09.2020, 17:00 Uhr, Math Machine Learning seminar MPI MIS + UCLA

• Randall Balestriero (Rice University)
• Max-Affine Spline Insights into Deep Networks
• see video

01.10.2020, 17:00 Uhr, Math Machine Learning seminar MPI MIS + UCLA

• David Rolnick (McGill University & Mila)
• Expressivity and learnability: linear regions in deep ReLU networks
• see video

08.10.2020, 17:00 Uhr, Math Machine Learning seminar MPI MIS + UCLA

• Yaoyu Zhang (Shanghai Jiao Tong University)
• Impact of initialization on generalization of deep neural networks
• see video

15.10.2020, 17:00 Uhr, Math Machine Learning seminar MPI MIS + UCLA

• Guy Blanc (Stanford University)
• Provable Guarantees for Decision Tree Induction
• see video

22.10.2020, 17:00 Uhr, Math Machine Learning seminar MPI MIS + UCLA

• Boris Hanin (Princeton University)
• Neural Networks at Finite Width and Large Depth
• see video

29.10.2020, 17:00 Uhr, Math Machine Learning seminar MPI MIS + UCLA

• Yaim Cooper (Institute for Advanced Study, Princeton)
• The geometry of the loss function of deep neural networks

05.11.2020, 11:00 Uhr, Deep Learning Theory Group Seminar

• Pierre Bréchet (MPI MIS, Leipzig)
• Wasserstein Gradient Flow (II): Mean-field approximation regime of Neural Networks

05.11.2020, 17:00 Uhr, Math Machine Learning seminar MPI MIS + UCLA

• Kenji Kawaguchi (MIT)
• Deep learning: theoretical results on optimization and mixup

12.11.2020, 17:00 Uhr, Math Machine Learning seminar MPI MIS + UCLA

• Yasaman Bahri (Google Brain)
• The Large Learning Rate Phase of Wide, Deep Neural Networks
• see video

17.11.2020, 18:00 Uhr, Deep Learning Theory Group Seminar

• Pierre Bréchet (MPI MIS, Leipzig)
• Wasserstein Gradient Flow (III): Mean-field approximation regime of Neural Networks

20.11.2020, 17:00 Uhr, Math Machine Learning seminar MPI MIS + UCLA

• Amartya Sanyal (University of Oxford)
• How benign is benign overfitting?
• see video

01.12.2020, 18:00 Uhr, Deep Learning Theory Group Seminar

• Johannes Müller (MPI MIS, Leipzig)
• Notions from RL: Overview of existing methods

15.12.2020, 18:00 Uhr, Deep Learning Theory Group Seminar

• Yuguang Wang (MPI MIS, Leipzig)
• Framelet graph convolution

17.12.2020, 17:00 Uhr, Math Machine Learning seminar MPI MIS + UCLA

• Tim G. J. Rudner (University of Oxford)
• Outcome-Driven Reinforcement Learning via Variational Inference