Finished

Thesis available for download here: Dinesh Parthsarathy: Safe Monte Carlo Tree Search using Learned Safety Critics Background Reinforcement Learning and planning are two research fields that have shown great success…

Recently, offline (sometimes also called ‘batch’) Reinforcement Learning (RL) algorithms have gained significant research traction [1]. The reason behind this is that – unlike in the classical Reinforcement Learning formulation…

Reinforcement Learning is broadly applicable for diverse tasks across many domains. On many problems, it has achieved superhuman performance [5]. However, the black-box neural networks used by modern RL algorithms…