Workshop IV: Efficient Tensor Representations for Learning and Computational Complexity - IPAM

Workshop IV: Efficient Tensor Representations for Learning and Computational Complexity

Part of the Long Program Tensor Methods and Emerging Applications to the Physical and Data Sciences
May 17 - 21, 2021

Schedule

All times in this Schedule are Pacific Time (PT)

Monday, May 17, 2021

Morning Session

07:55-07:55
SESSION CHAIR: Rong Ge (Duke Univ.)
07:55-08:00
Welcome & Opening Remarks: Dean Miguel García-Garibay (Dean of Physical Sciences, UCLA) and Dima Shlyakhtenko (Director, IPAM)
08:00-08:30
Anna Seigal (Harvard University)
Principal Components along Quiver Representations
08:40-08:55
Break
08:55-09:25
Christian Ikenmeyer (University of Liverpool)
Minrank: On the Complexity of Orbit Closures
09:35-09:50
Break
09:50-10:20
Visu Makam (Institute for Advanced Study)
Maximum Likelihood Estimation for Tensor Normal Models
10:30-10:45
Break
10:45-11:15
Ketan Mulmuley (University of Chicago)
Title not available
11:30-12:00
Poster Session: Kirandeep Kour, Deqing Wang and Eric Evert

Tuesday, May 18, 2021

Morning Session

08:00-08:00
SESSION CHAIR: Michael Walter (Univ. Amsterdam)
08:00-08:30
Alessandra Bernardi (Università di Trento)
On the Dimension of Tensor Network Varieties.
08:40-08:55
Break
08:55-09:25
Shmuel Friedland (University of Illinois at Chicago)
Complexity of Computation of Tensor Rank and Best Rank One Approximation
09:35-09:50
Break
09:50-10:20
Furong Huang (University of Maryland)
Understanding, Interpreting and Design Neural Network Models Through Tensor Representations
10:30-10:45
Break
10:45-11:15
Babak Hassibi (California Institute of Technology)
Implicit and Explicit Regularization in Deep Neural Networks
11:25-11:40
Break
11:40-12:10
Qing Qu (University of Michigan)
Landscape Analysis for Overcomplete Tensor and Neural Collapse

Wednesday, May 19, 2021

Morning Session

08:00-08:00
SESSION CHAIR: David Gross (Univ. Cologne)
08:00-08:30
Jean Kossaifi (Nvidia Corporation)
Efficient Tensor Representation for Deep Learning with TensorLy and PyTorch
08:40-08:55
Break
08:55-09:25
Matthias Christandl (University of Copenhagen)
Asymptotics of Ranks of Tensors
09:35-09:50
Break
09:50-10:20
Yuxin Chen (Princeton University)
The Effectiveness of Nonconvex Tensor Completion: Fast Convergence and Uncertainty Quantification.
10:30-10:45
Break
10:45-11:15
Rose Yu (University of California, San Diego (UCSD))
Multiresolution Tensor Learning for Efficient and Interpretable Spatial Analysis
11:25-11:40
Break
11:40-12:10
Jeroen Zuiddam (University of Amsterdam)
Subrank, partition rank and slice rank

Thursday, May 20, 2021

Morning Session

08:00-08:00
SESSION CHAIR: Matthias Christandl (Univ. Copenhagen)
08:00-08:30
Peter Burgisser (Technische Universität Berlin)
Towards a Theory of Non-commutative Optimization: Geodesic First and Second Order Methods for Moment Maps and Polytopes.
08:40-08:55
Break
08:55-09:25
Harm Derksen (Northeastern University)
Invariant Tensors and Wheeled PROPs.
09:35-09:50
Break
09:50-10:20
Daniel Hsu (Columbia University)
Computational Lower Bounds for Tensor PCA
10:30-10:45
Break
10:45-11:15
Andrea Montanari (Stanford University)
Simple Models for Noisy Tensors and Their Properties

Friday, May 21, 2021

Morning Session

08:00-08:00
SESSION CHAIR: Jianfeng Lu (Duke Univ.)
08:00-08:30
Dong Xia (Hong Kong University of Science and Technology)
Community Detection on Mixture Multi-layer Networks via Regularized Tensor Decomposition
08:40-08:55
Break
08:55-09:25
Richard Kueng (Johannes Kepler University)
Predicting Many Properties of a Quantum System From Very Few Measurements
09:35-09:50
Break
09:50-10:20
Holger Rauhut (RWTH Aachen University)
Learning Deep Matrix Factorizations Via Gradient Descent: Implicit Bias Towards Low Rank
10:30-10:45
Break
10:45-11:15
Hanie Sedghi (Google AI)
Taming Deep Convolutional Networks
Printable Version