Sanjeev Arora: Toward Theoretical Understanding of Deep Learning (ICML 2018 tutorial)
Audio starts at 1:46
Abstract:
We survey progress in recent years toward developing a theory of deep learning. Works have started addressing issues such as: (a) the effect of architecture choices on the optimization landscape, training speed, and expressiveness (b) quantifying the true "capacity" of the net, as a step towards understanding why nets with hugely more parameters than training examples nevertheless do not overfit (c) understanding inherent power and limitations of deep generative models, especially (various flavors of) generative adversarial nets (GANs) (d) understanding properties of simple RNN-style language models and some of their solutions (word embeddings and sentence embeddings)
While these are early results, they help illustrate what kind of theory may ultimately arise for deep learning.
Presented by Sanjeev Arorau (Princeton U., Inst. For Advanced Study)
The tutorial website: http://unsupervised.cs.princeton.edu/deeplearningtutorial.html
Видео Sanjeev Arora: Toward Theoretical Understanding of Deep Learning (ICML 2018 tutorial) канала Steven Van Vaerenbergh
Abstract:
We survey progress in recent years toward developing a theory of deep learning. Works have started addressing issues such as: (a) the effect of architecture choices on the optimization landscape, training speed, and expressiveness (b) quantifying the true "capacity" of the net, as a step towards understanding why nets with hugely more parameters than training examples nevertheless do not overfit (c) understanding inherent power and limitations of deep generative models, especially (various flavors of) generative adversarial nets (GANs) (d) understanding properties of simple RNN-style language models and some of their solutions (word embeddings and sentence embeddings)
While these are early results, they help illustrate what kind of theory may ultimately arise for deep learning.
Presented by Sanjeev Arorau (Princeton U., Inst. For Advanced Study)
The tutorial website: http://unsupervised.cs.princeton.edu/deeplearningtutorial.html
Видео Sanjeev Arora: Toward Theoretical Understanding of Deep Learning (ICML 2018 tutorial) канала Steven Van Vaerenbergh
Показать
Комментарии отсутствуют
Информация о видео
Другие видео канала
![Deep Learning: A Crash Course](https://i.ytimg.com/vi/r0Ogt-q956I/default.jpg)
![Benjamin Recht: Optimization Perspectives on Learning to Control (ICML 2018 tutorial)](https://i.ytimg.com/vi/nF2-39a29Pw/default.jpg)
![The Heidelberg Laureate Forum Foundation presents the HLF Portraits: Sanjeev Arora](https://i.ytimg.com/vi/AFrQNcG0Qbo/default.jpg)
![PyTorch at Tesla - Andrej Karpathy, Tesla](https://i.ytimg.com/vi/oBklltKXtDE/default.jpg)
![The science of cells that never get old | Elizabeth Blackburn](https://i.ytimg.com/vi/2wseM6wWd74/default.jpg)
![AAAI 20 / AAAI 2020 Keynotes Turing Award Winners Event / Geoff Hinton, Yann Le Cunn, Yoshua Bengio](https://i.ytimg.com/vi/UX8OubxsY8w/default.jpg)
![Ian Goodfellow: Adversarial Machine Learning (ICLR 2019 invited talk)](https://i.ytimg.com/vi/sucqskXRkss/default.jpg)
![Sanjeev Arora: Why do deep nets generalize, that is, predict well on unseen data](https://i.ytimg.com/vi/xscvWCC-y6U/default.jpg)
![Model-Based Policy Optimization (ICML Workshops)](https://i.ytimg.com/vi/rdF7q8MipRs/default.jpg)
![How Machines Learn](https://i.ytimg.com/vi/R9OHn5ZF4Uo/default.jpg)
![ICLR Debate with Leslie Kaelbling (ICLR 2019)](https://i.ytimg.com/vi/veG8S5rqKIE/default.jpg)
![IDSS Distinguished Speaker Seminar: Sanjeev Arora | March 5, 2019](https://i.ytimg.com/vi/N-ttnuDMJwg/default.jpg)
![A friendly introduction to Convolutional Neural Networks and Image Recognition](https://i.ytimg.com/vi/2-Ol7ZB0MmU/default.jpg)
![Variational Inference: Foundations and Modern Methods (NIPS 2016 tutorial)](https://i.ytimg.com/vi/ogdv_6dbvVQ/default.jpg)
![How AI can save our humanity | Kai-Fu Lee](https://i.ytimg.com/vi/ajGgd9Ld-Wc/default.jpg)
![Deep Learning Basics: Introduction and Overview](https://i.ytimg.com/vi/O5xeyoRL95U/default.jpg)
![Complete Statistical Theory of Learning (Vladimir Vapnik) | MIT Deep Learning Series](https://i.ytimg.com/vi/Ow25mjFjSmg/default.jpg)
![Sanjeev Arora - Is Optimization the Right Language to Understand Deep Learning?](https://i.ytimg.com/vi/HMdJd2minAI/default.jpg)
![Google's Deep Mind Explained! - Self Learning A.I.](https://i.ytimg.com/vi/TnUYcTuZJpM/default.jpg)
![“Learning to Code is Not Just for Coders” | Ali Partovi | TEDxSausalito](https://i.ytimg.com/vi/MvTSPwftvyo/default.jpg)