We study a kind of new SDE that was arisen from the research on optimiza...
Neural networks have shown great potential in accelerating the solution ...
Speech emotion recognition (SER) classifies audio into emotion categorie...
Stochastic partial differential equations (SPDEs) are significant tools ...
Modeling many-body systems has been a long-standing challenge in science...
The momentum acceleration technique is widely adopted in many optimizati...
Dropout is a powerful and widely used technique to regularize the traini...
Denoising diffusion probabilistic models have been recently proposed to
...
Learning dynamics governed by differential equations is crucial for
pred...
Energy conservation is a basic physics principle, the breakdown of which...
Transformer architecture achieves great success in abundant natural lang...
Batch normalization (BN) has become a crucial component across diverse d...
Despite their overwhelming capacity to overfit, deep neural networks tra...
The way the basis path set works in neural network remains mysterious, a...
Stochastic gradient descent (SGD) and its variants are mainstream method...
Based on basis path set, G-SGD algorithm significantly outperforms
conve...
Value function estimation is an important task in reinforcement learning...
It was empirically confirmed by Keskar et al.SharpMinima that flatter
mi...
Q-learning is one of the most popular methods in Reinforcement Learning ...
Recently, path norm was proposed as a new capacity measure for neural
ne...
Asynchronous stochastic gradient descent (ASGD) is a popular parallel
op...
It has been widely observed that many activation functions and pooling
m...
When using stochastic gradient descent to solve large-scale machine lear...
Many machine learning tasks can be formulated as Regularized Empirical R...