Country
Chefs'RandomTables: Non-TrigonometricRandom Features
We introduce chefs' random tables(CRTs), a new class of non-trigonometric random features (RFs) toapproximate Gaussian andsoftmax-kernels.
Learning and Generalization in Overparameterized Neural Networks, Going Beyond Two Layers
Zeyuan Allen-Zhu, Yuanzhi Li, Yingyu Liang
The fundamental learning theory behind neural networks remains largely open. What classes of functions can neural networks actually learn?
Practical Equivariances via Relational Conditional Neural Processes Daolang Huang 1 Manuel Haussmann
However, prior attempts to include equivariances in CNPs do not scale effectively beyond two input dimensions.
Gradient-basedAdaptiveMarkovChainMonteCarlo
The basic set up of an MCMC algorithm inanyprobabilistic (e.g.
Dendritic cortical microcircuits approximate the backpropagation algorithm
João Sacramento, Rui Ponte Costa, Yoshua Bengio, Walter Senn
Neural Information Processing Systems http://nips.cc/
Regularized Gradient Descent Ascentfor Two-Player Zero-Sum Markov Games
Under >0, ( )is 2 SB, and ( )isunique 2 SA.
