Zobrazeno 1 - 10
of 1 692
pro vyhledávání: '"Patterson, Andrew"'
Autor:
Ivanov, Aleksei V., Patterson, Andrew, Bothe, Marius, Sünderhauf, Christoph, Berntson, Bjorn K., Mortensen, Jens Jørgen, Kuisma, Mikael, Campbell, Earl, Izsák, Róbert
Quantum simulation of materials is a promising application area of quantum computers. In order to realize this promise, finding ways to reduce quantum resources while maintaining the accuracy of results will be necessary. In electronic structure calc
Externí odkaz:
http://arxiv.org/abs/2408.03159
This paper introduces a new empirical methodology, the Cross-environment Hyperparameter Setting Benchmark, that compares RL algorithms across environments using a single hyperparameter setting, encouraging algorithmic development which is insensitive
Externí odkaz:
http://arxiv.org/abs/2407.18840
In this work, we present composite Bernstein polynomials as a direct collocation method for approximating optimal control problems. An analysis of the convergence properties of composite Bernstein polynomials is provided, and beneficial properties of
Externí odkaz:
http://arxiv.org/abs/2407.18081
Experience replay is ubiquitous in reinforcement learning, to reuse past data and improve sample efficiency. Though a variety of smart sampling schemes have been introduced to improve performance, uniform sampling by far remains the most common appro
Externí odkaz:
http://arxiv.org/abs/2407.09702
Offline reinforcement learning algorithms often require careful hyperparameter tuning. Consequently, before deployment, we need to select amongst a set of candidate policies. As yet, however, there is little understanding about the fundamental limits
Externí odkaz:
http://arxiv.org/abs/2312.02355
Empirical design in reinforcement learning is no small task. Running good experiments requires attention to detail and at times significant computational resources. While compute resources available per dollar have continued to grow rapidly, so have
Externí odkaz:
http://arxiv.org/abs/2304.01315
The approximate state estimation and the closely related classical shadows methods allow for the estimation of complicated observables with relatively few shots. As these methods make use of random measurements that can symmetrise the effect of reado
Externí odkaz:
http://arxiv.org/abs/2303.17741
Autor:
Yuan Tian1,2, Rimal, Bipin1, Bisanz, Jordan E.3,4, Wei Gui2, Wolfe, Trenton M.5, Imhoi Koo1, Murray, Iain A.1, Nettleford, Shaneice K.1, Shigetoshi Yokoyama1, Fangcong Dong1, Sergei Koshkin2, Prabhu, K. Sandeep1, Turnbaugh, Peter J.4,6, Walk, Seth T.5, Perdew, Gary H.1, Patterson, Andrew D.1 adp117@psu.edu
Publikováno v:
Environmental Health Perspectives. Aug2024, Vol. 132 Issue 8, p087005-1-087005-14. 14p.
Most value function learning algorithms in reinforcement learning are based on the mean squared (projected) Bellman error. However, squared errors are known to be sensitive to outliers, both skewing the solution of the objective and resulting in high
Externí odkaz:
http://arxiv.org/abs/2205.08464