no code implementations • 23 Jun 2023 • Jamie F. Mair, Luke Causer, Juan P. Garrahan

Most iterative neural network training methods use estimates of the loss function over small random subsets (or minibatches) of the data to update the parameters, which aid in decoupling the training time from the (often very large) size of the training datasets.

1 code implementation • 28 Sep 2022 • Edward Gillman, Dominic C. Rose, Juan P. Garrahan

We present a framework to integrate tensor network (TN) methods with reinforcement learning (RL) for solving dynamical optimisation tasks.

no code implementations • 22 Sep 2022 • Jamie F. Mair, Dominic C. Rose, Juan P. Garrahan

In machine learning, there is renewed interest in neural network ensembles (NNEs), whereby predictions are obtained as an aggregate from a diverse set of smaller models, rather than from a single larger model.

no code implementations • 10 May 2021 • Avishek Das, Dominic C. Rose, Juan P. Garrahan, David T. Limmer

We present a method to probe rare molecular dynamics trajectories directly using reinforcement learning.

no code implementations • 11 Mar 2021 • Stefano Marcantoni, Carlos Pérez-Espigares, Juan P. Garrahan

We derive a general scheme to obtain quantum fluctuation relations for dynamical observables in open quantum systems.

Quantum Physics Statistical Mechanics

1 code implementation • 26 May 2020 • Dominic C. Rose, Jamie F. Mair, Juan P. Garrahan

By minimising the distance between a reweighted ensemble and that of a suitably parametrised controlled dynamics we arrive at a set of methods similar to those of RL to numerically approximate the optimal dynamics that realises the rare behaviour of interest.

no code implementations • 12 Feb 2020 • Edward Gillman, Dominic C. Rose, Juan P. Garrahan

Tensor network (TN) techniques - often used in the context of quantum many-body physics - have shown promise as a tool for tackling machine learning (ML) problems.

Cannot find the paper you are looking for? You can
Submit a new open access paper.

Contact us on:
hello@paperswithcode.com
.
Papers With Code is a free resource with all data licensed under CC-BY-SA.