Gintare Karolina Dziugaite - Distribution-dependent generalization bounds

preview_player
Показать описание
Guest talk by Gintare Karolina Dziugaite on "Distribution-dependent generalization bounds for noisy, iterative learning algorithms"

This talk is part of the seminar series held by MTL MLOpt:

Host: Ioannis Mitliagkas
Organization: Reyhane Askari Hemmat, Ioannis Mitliagkas

October 2020, Montréal.
  1. MTL MLOpt
Рекомендации по теме
Gintare Karolina Dziugaite 1:05:18

Gintare Karolina Dziugaite - Distribution-dependent generalization bounds

Gintare Karolina Dziugaite 0:51:37

Gintare Karolina Dziugaite - The dynamics of memorization and generalization in deep learning

Progress towards understanding 1:01:26

Progress towards understanding generalization in Deep Learning | Gintarė Karolina Džiugaitė

Gintarė Karolina Džiugaitė, 1:04:29

Gintarė Karolina Džiugaitė, Towards explaining generalization for interpolating predictors

PAC-Bayesian approaches to 0:31:34

PAC-Bayesian approaches to understanding generalization in deep learning - Gintare Dziugaite

IMG_0706 0:00:38

IMG_0706

Karolina Dziugaite on 0:26:02

Karolina Dziugaite on Nonvacuous Generalization Bounds for Deep Neural Networks via PAC-Bayes

Lecture 8: Progress 1:02:26

Lecture 8: Progress Towards Understanding Generalization in Deep Learning (English)

Pokalbiai apie Dirbtinį 0:27:57

Pokalbiai apie Dirbtinį Intelektą su Gintare Karolina Džiugaite

IMG_0711 0:01:37

IMG_0711

Studying Generalization in 0:44:40

Studying Generalization in Deep Learning via PAC-Bayes

IMG_0707 0:01:35

IMG_0707

MLT __init__ Session 0:35:45

MLT __init__ Session #15: Deep Learning on a Data Diet

Tightening information-theoretic generalization 0:28:11

Tightening information-theoretic generalization bounds with data-dependent estimate... - Daniel Roy

Size-free Generalization Bounds 0:34:34

Size-free Generalization Bounds for Convolutional Neural Networks

PaperView: 'Deep learning 0:21:34

PaperView: 'Deep learning on a data diet'

Distribution-Induced Bidirectional Generative 0:01:01

Distribution-Induced Bidirectional Generative Adversarial Network for Graph Representation Learning

When and How 0:05:53

When and How CNNs Generalize to Out-of-Distribution Category-Viewpoint Combinations

Lecture 9: Information-Theoretic 1:05:38

Lecture 9: Information-Theoretic Generalization Bounds for Stochastic Gradient Descent (English)

Princeton ORFE Deep 5:50:05

Princeton ORFE Deep Learning Theory Summer School -- Day 3

Tighter risk certificates 0:56:53

Tighter risk certificates for (probabilistic) neural networks

Carl Vondrick | 0:37:59

Carl Vondrick | Data and Task Generalization | PTSGM@ECCV2020

Resilient Representation and 0:42:18

Resilient Representation and Provable Generalization

Dichotomize and Generalize: 0:09:52

Dichotomize and Generalize: PAC-Bayesian Binary Activated Deep Neural Networks

Комментарии
Автор

Please, I would like to know if there are papers or books that summarise the main advances in generalisations bounds so I can find gaps in literature to work on.

anas.k