Phillip Isola -- When and Why Does Contrastive Learing Work?

  Рет қаралды 6,667

Learning with Limited and Imperfect Data

Learning with Limited and Imperfect Data

Күн бұрын

Пікірлер: 2
@420_gunna
@420_gunna Жыл бұрын
awesome
@farrukhrahman6779
@farrukhrahman6779 3 жыл бұрын
Great Talk! The idea that aligned and uniformity in representation space generalize across tasks & domains seems pretty powerful, especially when we do not know our downstream task a priori. Can we perhaps use these metrics to dynamically explore the multiview/augmentation space (ie. use the metrics to guide something like AutoAugment during the contrastive learning phase)?
Contrastive Learning in PyTorch - Part 1: Introduction
14:21
DeepFindr
Рет қаралды 37 М.
When Does Contrastive Visual Representation Learning Work? - CVPR 2022
4:17
Caltech Computational Vision Lab
Рет қаралды 2,1 М.
Мен атып көрмегенмін ! | Qalam | 5 серия
25:41
Что-что Мурсдей говорит? 💭 #симбочка #симба #мурсдей
00:19
IL'HAN - Qalqam | Official Music Video
03:17
Ilhan Ihsanov
Рет қаралды 700 М.
Contrastive Predictive Coding (CPC)
17:11
Deep Foundations
Рет қаралды 6 М.
Can Contrastive Learning Work? -  SimCLR Explained
9:35
Boris Meinardus
Рет қаралды 12 М.
Why Does Diffusion Work Better than Auto-Regression?
20:18
Algorithmic Simplicity
Рет қаралды 446 М.
Why Does Batch Norm Work? (C2W3L06)
11:40
DeepLearningAI
Рет қаралды 203 М.
What Are Neural Networks Even Doing? (Manifold Hypothesis)
13:20
Ishan Misra: General purpose visual recognition across modalities with limited supervision
25:44
Learning with Limited and Imperfect Data
Рет қаралды 850
Contrastive Learning - 5 Minutes with Cyrill
5:24
Cyrill Stachniss
Рет қаралды 20 М.
Мен атып көрмегенмін ! | Qalam | 5 серия
25:41