A deep reinforcement learning agent can relax some Al-Fe crystals in fewer steps than BFGS or conjugate gradient, but the main benchmark is biased because the agent is trained on the test structures.
A low resources space time approach to the GW approximation
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cond-mat.mtrl-sci 1years
2025 1verdicts
REJECT 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Acceleration of crystal structure relaxation with Deep Reinforcement Learning
A deep reinforcement learning agent can relax some Al-Fe crystals in fewer steps than BFGS or conjugate gradient, but the main benchmark is biased because the agent is trained on the test structures.