Tail worst-case latency for multi-robot graph monitoring is exactly reformulated as an average-reward event-driven MDP (TWLO-MDP), enabling RL methods that beat baselines on synthetic and crime-map graphs.
Learning Montezuma's Revenge from a Single Demonstration
4 Pith papers cite this work, alongside 84 external citations. Polarity classification is still indexing.
4
Pith papers citing it
84
external citations · Pith
abstract
We propose a new method for learning from a single demonstration to solve hard exploration tasks like the Atari game Montezuma's Revenge. Instead of imitating human demonstrations, as proposed in other recent works, our approach is to maximize rewards directly. Our agent is trained using off-the-shelf reinforcement learning, but starts every episode by resetting to a state from a demonstration. By starting from such demonstration states, the agent requires much less exploration to learn a game compared to when it starts from the beginning of the game at every episode. We analyze reinforcement learning for tasks with sparse rewards in a simple toy environment, where we show that the run-time of standard RL methods scales exponentially in the number of states between rewards. Our method reduces this to quadratic scaling, opening up many tasks that were previously infeasible. We then apply our method to Montezuma's Revenge, for which we present a trained agent achieving a high-score of 74,500, better than any previously published result.
representative citing papers
LLMs exhibit Bayesian-like hypothesis updating with strong-sampling bias and an evaluation-generation gap but generalize poorly outside observed data.
ULPS integrates A*-generated symbolic trajectories, fine-tuned BERT priors, MC dropout uncertainty, and entropy-based blending into PPO, reporting over 9% accuracy gains and better sample efficiency on the MiniGridUnlockPickup benchmark.
A curriculum sampling questions with high variance in success rate improves reinforcement learning performance for LLM reasoning tasks.
citing papers explorer
-
Minimizing Worst-Case Weighted Latency for Multi-Robot Persistent Monitoring: Theory and RL-Based Solutions
Tail worst-case latency for multi-robot graph monitoring is exactly reformulated as an average-reward event-driven MDP (TWLO-MDP), enabling RL methods that beat baselines on synthetic and crime-map graphs.
-
Hypothesis generation and updating in large language models
LLMs exhibit Bayesian-like hypothesis updating with strong-sampling bias and an evaluation-generation gap but generalize poorly outside observed data.
-
Uncertainty-Aware LLM-Guided Policy Shaping for Sparse-Reward Reinforcement Learning
ULPS integrates A*-generated symbolic trajectories, fine-tuned BERT priors, MC dropout uncertainty, and entropy-based blending into PPO, reporting over 9% accuracy gains and better sample efficiency on the MiniGridUnlockPickup benchmark.
-
Learning to Reason at the Frontier of Learnability
A curriculum sampling questions with high variance in success rate improves reinforcement learning performance for LLM reasoning tasks.