Workshop
Heavy Tails in ML: Structure, Stability, Dynamics
Mert Gurbuzbalaban 路 Stefanie Jegelka 路 Michael Mahoney 路 Umut Simsekli
Room R02-R05 (level 2)
Fri 15 Dec, 7 a.m. PST
Heavy-tails and chaotic behavior naturally appear in many ways in ML. We aim to understand how they emerge and how they affect the properties of ML methods.
Chat is not available.
Timezone: America/Los_Angeles
Schedule
|
Fri 7:00 a.m. - 7:01 a.m.
|
Opening Remarks
(
Presentation
)
>
|
馃敆 |
|
Fri 7:00 a.m. - 8:00 a.m.
|
Keynote: Adam Wierman
(
Presentation
)
>
SlidesLive Video |
馃敆 |
|
Fri 8:00 a.m. - 8:30 a.m.
|
Coffee break
|
馃敆 |
|
Fri 8:30 a.m. - 9:10 a.m.
|
Invited talk: Liam Hodgkinson
(
Presentation
)
>
SlidesLive Video |
馃敆 |
|
Fri 9:10 a.m. - 9:35 a.m.
|
Contributed talk: Vivien Cabannes
(
Presentation
)
>
SlidesLive Video |
馃敆 |
|
Fri 9:35 a.m. - 10:00 a.m.
|
Contributed talk: Dominique Perrault-Joncas
(
Presentation
)
>
SlidesLive Video |
馃敆 |
|
Fri 10:00 a.m. - 12:00 p.m.
|
Lunch break
|
馃敆 |
|
Fri 12:00 p.m. - 12:40 p.m.
|
Invited talk: Nisha Chandramoorthy
(
Presentation
)
>
SlidesLive Video |
馃敆 |
|
Fri 12:40 p.m. - 1:05 p.m.
|
Contributed talk: Jeremy Cohen
(
Presentation
)
>
SlidesLive Video |
馃敆 |
|
Fri 1:05 p.m. - 1:30 p.m.
|
Coffee break
|
馃敆 |
|
Fri 1:30 p.m. - 2:10 p.m.
|
Invited talk: Charles Martin
(
Presentation
)
>
SlidesLive Video |
馃敆 |
|
Fri 2:10 p.m. - 3:30 p.m.
|
Poster session
(
Poster presentation
)
>
SlidesLive Video |
馃敆 |
|
Fri 3:29 p.m. - 3:30 p.m.
|
Closing remarks
(
Presentation
)
>
|
馃敆 |
|
-
|
Semi-Implicit Neural Ordinary Differential Equations for Learning Chaotic Systems ( Poster ) > link | Hong Zhang 路 Ying Liu 路 Romit Maulik 馃敆 |
|
-
|
From Mutual Information to Expected Dynamics: New Generalization Bounds for Heavy-Tailed SGD ( Poster ) > link | Benjamin, Bernard, Louis DUPUIS 路 Paul Viallard 馃敆 |
|
-
|
Instance-Aware Repeat Factor Sampling for Long-Tailed Object Detection ( Poster ) > link | Burhan Yaman 路 Tanvir Mahmud 路 Chun-Hao Liu 馃敆 |
|
-
|
Meta-Analysis of Randomized Experiments with Applications to Heavy-Tailed Response Data ( Poster ) > link | Nilesh Tripuraneni 路 Dominique Perrault-Joncas 路 Dhruv Madeka 路 Dean Foster 路 Michael Jordan 馃敆 |
|
-
|
The Effects of Ensembling on Long-Tailed Data ( Poster ) > link | Estefany Kelly Buchanan 路 Geoff Pleiss 路 John Cunningham 馃敆 |
|
-
|
High-dimensional robust regression under heavy-tailed data: Asymptotics and Universality ( Poster ) > link | Urte Adomaityte 路 Leonardo Defilippis 路 Bruno Loureiro 路 Gabriele Sicuro 馃敆 |
|
-
|
Large Deviations and Metastability Analysis for Heavy-Tailed Dynamical Systems ( Poster ) > link | Chang-Han Rhee 路 Xingyu Wang 馃敆 |
|
-
|
Associative Memories with Heavy-Tailed Data ( Poster ) > link | Vivien Cabannes 路 Elvis Dohmatob 路 Alberto Bietti 馃敆 |
|
-
|
Outliers with Opposing Signals Have an Outsized Effect on Neural Network Optimization ( Poster ) > link | Elan Rosenfeld 路 Andrej Risteski 馃敆 |
|
-
|
Deep neural networks with dependent weights: \\Gaussian Process mixture limit, heavy tails, sparsity and compressibility ( Poster ) > link | Hoil Lee 路 Fadhel Ayed 路 Paul Jung 路 Juho Lee 路 Hongseok Yang 路 Francois Caron 馃敆 |
|
-
|
Adaptive Gradient Methods at the Edge of Stability ( Poster ) > link | Jeremy M Cohen 路 Behrooz Ghorbani 路 Shankar Krishnan 路 Naman Agarwal 路 Sourabh Medapati 路 Michal Badura 路 Daniel Suo 路 Zachary Nado 路 George Dahl 路 Justin Gilmer 馃敆 |
|
-
|
Online Student- Processes with an Overall-local Scale Structure for Modelling Non-stationary Data ( Poster ) > link | Taole Sha 路 Michael Minyi Zhang 馃敆 |
|
-
|
On the Varied Faces of Overparameterization in Supervised and Self-Supervised Learning ( Poster ) > link | Matteo Gamba 路 Arna Ghosh 路 Kumar Krishna Agrawal 路 Blake Richards 路 Hossein Azizpour 路 M氓rten Bj枚rkman 馃敆 |
|
-
|
Neural network compression with heavy-tailed SGD ( Poster ) > link | Yijun Wan 路 Abdellatif Zaidi 路 Umut Simsekli 馃敆 |
|
-
|
Representation Learning for Extremes ( Poster ) > link | Ali Hasan 路 Yuting Ng 路 Jose Blanchet 路 Vahid Tarokh 馃敆 |
|
-
|
High Probability Guarantees for Random Reshuffling ( Poster ) > link | Hengxu Yu 路 Xiao Li 馃敆 |
|
-
|
Towards Fully Adaptive Regret Minimization in Heavy-Tailed Bandits ( Poster ) > link | Gianmarco Genalti 路 Lupo Marsigli 路 Nicola Gatti 路 Alberto Maria Metelli 馃敆 |
|
-
|
Robust gradient estimation in the presence of heavy-tailed noise ( Poster ) > link | Fabian Schaipp 路 Umut Simsekli 路 Robert Gower 馃敆 |
|
-
|
Generalised Hyperbolic State-space Models for Inference in Dynamic Systems ( Poster ) > link | Yaman Kindap 路 Simon Godsill 馃敆 |
|
-
|
Variational Inference for SDEs Driven by Fractional Noise ( Poster ) > link | Rembert Daems 路 Manfred Opper 路 Guillaume Crevecoeur 路 Tolga Birdal 馃敆 |
|
-
|
Revisiting the noise Model of SGD ( Poster ) > link | Barak Battash 路 Ofir Lindenbaum 馃敆 |