most citedReborn Mechanism: Rethinking the Negative Phase Information Flow in Convolutional Neural Network

2 citations · 5 across the 8 of their papers we have counts for

collaborators
Showing cs.LGShow all

3 papers · 1 filter

cs.LG2023

Evolution: A Unified Formula for Feature Operators from a High-level Perspective

Zhicheng Cai

Traditionally, different types of feature operators (e.g., convolution, self-attention and involution) utilize different approaches to extract and aggregate the features. Resemblan…

cs.LG2021★ 2 cited

SA-GD: Improved Gradient Descent Learning Strategy with Simulated Annealing

Zhicheng Cai

Gradient descent algorithm is the most utilized method when optimizing machine learning issues. However, there exists many local minimums and saddle points in the loss function, es…

cs.LG2021

Jitter: Random Jittering Loss Function

Zhicheng Cai, Chenglei Peng, Sidan Du

Regularization plays a vital role in machine learning optimization. One novel regularization method called flooding makes the training loss fluctuate around the flooding level. It…