This paper explores the imperative need and methodology for developing a...
We present a simple framework for one-class classification and anomaly
d...
Pruning neural networks before training has received increasing interest...
The past decade has witnessed a drastic increase in modern deep neural
n...
Neural networks that satisfy invariance with respect to input permutatio...
Deep neural networks are vulnerable to adversarial attacks. Ideally, a r...
Generative adversarial nets (GANs) have been remarkably successful at
le...
Modern neural networks are often quite wide, causing large memory and
co...
In adversarial machine learning, deep neural networks can fit the advers...
Ever since Reddi et al. 2018 pointed out the divergence issue of Adam, m...
Model-agnostic meta-learning (MAML) and its variants have become popular...
Differential privacy (DP) is an essential technique for privacy-preservi...
Many existing federated learning (FL) algorithms are designed for superv...
The momentum acceleration technique is widely adopted in many optimizati...
Recent theoretical works on over-parameterized neural nets have focused ...
The Barzilai-Borwein (BB) method has demonstrated great empirical succes...
Understanding of GAN training is still very limited. One major challenge...
Recent advances in unsupervised representation learning have experienced...
Nonconvex-concave min-max problem arises in many machine learning
applic...
Network pruning, or sparse network has a long history and practical
sign...
One of the major concerns for neural network training is that the
non-co...
Gradient-based meta-learning (GBML) with deep neural nets (DNNs) has bec...
Current state-of-the-art object detectors are at the expense of high
com...
In distributed optimization, a popular technique to reduce communication...
Traditional landscape analysis of deep neural networks aims to show that...
When and why can a neural network be successfully trained? This article
...
Does over-parameterization eliminate sub-optimal local minima for neural...
It was recently found that the standard version of multi-block cyclic AD...
Terrain traversability analysis is a fundamental issue to achieve the
au...
Generative adversarial nets (GANs) and variational auto-encoders have
si...
In this paper, we study the loss surface of the over-parameterized fully...
This paper studies a class of adaptive gradient based momentum algorithm...
One of the main difficulties in analyzing neural networks is the
non-con...
It is widely conjectured that the reason that training algorithms for ne...
While Truncated Back-Propagation through Time (BPTT) is the most popular...
The joint base station (BS) association and beamforming problem has been...