CLARIPHY Topical Meeting - Dynamical mean-field theory for SGD in non-linear neural networks

preview_player
Показать описание
Video from CLARIPHY Topical Meeting - 21 Jan 2021

Slides/agenda:

Time markers for the talks:
00:00-40:20 Dynamical mean-field theory for Stochastic Gradient Descent in non-linear neural networks (Francesca Mignacco, Institut de Physique Théorique, CEA, Saclay)
  1. CLARIPHY
  2. CLARIPHY
  3. HEP
  4. DUNE
  5. XENON
  6. IceCube
Рекомендации по теме
CLARIPHY Topical Meeting 0:40:21

CLARIPHY Topical Meeting - Dynamical mean-field theory for SGD in non-linear neural networks

CLARIPHY Topical Meeting 1:00:33

CLARIPHY Topical Meeting (6 Jan 2021) - Interfacing theoretical parameter spaces and predictions

CLARIPHY Topical Meeting 0:55:09

CLARIPHY Topical Meeting (13 Jan 2021) - Learning Physical Interaction for Scene Understanding

CLARIPHY Topical Meeting 1:19:04

CLARIPHY Topical Meeting (04 Dec 2020) - Graph Neural Networks for Reconstruction

CLARIPHY Topical Meeting 1:01:20

CLARIPHY Topical Meeting (10 Mar 2021) - Learning to accelerate the simulation of PDEs

Dynamical mean-field theory 2:52:28

Dynamical mean-field theory

DACBench: Benchmarking Dynamic 0:13:46

DACBench: Benchmarking Dynamic Algorithm Configuration | IJCAI 2021

IRIS-HEP Topical Meeting 0:49:25

IRIS-HEP Topical Meeting (21 Oct 2020) - GNNs for event reconstruction

Francesa Mignacco - 0:23:33

Francesa Mignacco - Dynamical Mean-Field Theory for SGD in Gaussian Mixture Classification

Vladimir Lobaskin: SmartNanoTox 0:25:33

Vladimir Lobaskin: SmartNanoTox International Conference Session 4 “Databases/Data management”.

Understanding the Dynamics 0:20:24

Understanding the Dynamics of Multi-pass SGD in High Dimensional Non-convex Problems

#31 Dynamical mean 0:59:58

#31 Dynamical mean field theory and Anderson Impurity model

Explorations at the 1:25:41

Explorations at the Physics ∩ ML Interface

Data Learning - 0:59:03

Data Learning - Statistical physics of stochastic gradient descent

09 Machine learning 0:31:21

09 Machine learning of physics theories and its universal tradeoff between accuracy and simplicity

Simulation-based inference for 0:45:43

Simulation-based inference for neuroscience (and beyond)

MSML2020 Invited Talk 0:43:33

MSML2020 Invited Talk by Prof. Lenka Zdeborová, CNRS France

Data Learning - 0:59:36

Data Learning - The frontier of Simulation-Based Inference

iMat - Digitalizing, 0:03:07

iMat - Digitalizing, democratising and empowering materials development via Artificial Intelligence

TestDive 2019 Conference: 0:41:10

TestDive 2019 Conference: Karol Przystalski - 'Machine learning in test automation'

Gabi Kotliar - 2:03:53

Gabi Kotliar - Intro to Dynamical Mean Field Theory and Electronic Structure of Correlated Materials

Kyle Cranmer - 0:48:39

Kyle Cranmer - Simulation-based Inference for Gravitational Wave Astronomy - IPAM at UCLA

Lenka Zdeborová, 'What 1:00:05

Lenka Zdeborová, 'What physics teaches us about computation in high dimensions'

Clariphy | Premios 0:00:55

Clariphy | Premios Sadosky 2015