research
∙
09/04/2023
Gated recurrent neural networks discover attention
Recent architectural developments have enabled recurrent neural networks...
share
research
∙
09/15/2022
Random initialisations performing above chance and how to find them
Neural networks trained with stochastic gradient descent (SGD) starting ...
research
∙
01/11/2021
Evaluating Disentanglement of Structured Latent Representations
We design the first multi-layer disentanglement metric operating at all ...
research
∙
12/04/2020
An O(n) time algorithm for finding Hamilton cycles with high probability
We design a randomized algorithm that finds a Hamilton cycle in 𝒪(n) tim...
research
∙
02/12/2020
An Optimal Decentralized (Δ+ 1)-Coloring Algorithm
Consider the following simple coloring algorithm for a graph on n vertic...
research
∙
10/11/2019
Improving Gradient Estimation in Evolutionary Strategies With Past Descent Directions
Evolutionary Strategies (ES) are known to be an effective black-box opti...
research
∙
10/11/2019
Decoupling Hierarchical Recurrent Neural Networks With Locally Computable Losses
Learning long-term dependencies is a key long-standing challenge of recu...
research
∙
02/11/2019
Optimal Kronecker-Sum Approximation of Real Time Recurrent Learning
One of the central goals of Recurrent Neural Networks (RNNs) is to learn...
research
∙
08/16/2018
The linear hidden subset problem for the (1+1) EA with scheduled and adaptive mutation rates
We study unbiased (1+1) evolutionary algorithms on linear functions with...
research
∙
08/03/2018
When Does Hillclimbing Fail on Monotone Functions: An entropy compression argument
Hillclimbing is an essential part of any optimization algorithm. An impo...
research
∙
05/28/2018
Approximating Real-Time Recurrent Learning with Random Kronecker Factors
Despite all the impressive advances of recurrent neural networks, sequen...
research
∙
01/22/2018
Even flying cops should think ahead
We study the entanglement game, which is a version of cops and robbers, ...
research
∙
05/24/2017
Fast-Slow Recurrent Neural Networks
Processing sequential data of variable length is a major challenge in a ...
research
∙
08/10/2016