Statistical physics of deep neural networks: Initialization toward optimal channels
In deep learning, neural networks serve as noisy channels between input data and its latent representation. This perspective naturally relates deep learning with the pursuit of constructing channels with optimal performance in information transmission and representation. While considerable efforts a...
Main Authors: | Kangyu Weng, Aohua Cheng, Ziyang Zhang, Pei Sun, Yang Tian |
---|---|
Format: | Article |
Language: | English |
Published: |
American Physical Society
2023-04-01
|
Series: | Physical Review Research |
Online Access: | http://doi.org/10.1103/PhysRevResearch.5.023023 |
Similar Items
-
Gradient Statistics-Based Multi-Objective Optimization in Physics-Informed Neural Networks
by: Sai Karthikeya Vemuri, et al.
Published: (2023-10-01) -
DeepMCGCN: Multi-channel Deep Graph Neural Networks
by: Lei Meng, et al.
Published: (2024-02-01) -
Sequence Recommendation with Dual Channel Heterogeneous Graph Neural Network
by: WU Jinchen, YANG Xingyao, YU Jiong, LI Ziyang, HUANG Shanhang, SUN Xinjie
Published: (2023-06-01) -
Multivariate Process Control Chart Pattern Classification Using Multi-Channel Deep Convolutional Neural Networks
by: Chuen-Sheng Cheng, et al.
Published: (2023-07-01) -
Deep Imitation Learning for Optimal Trajectory Planning and Initial Condition Optimization for an Unstable Dynamic System
by: Bo-Hsun Chen, et al.
Published: (2024-01-01)