SOTAVerified

Montezuma's Revenge

Montezuma's Revenge is an ATARI 2600 Benchmark game that is known to be difficult to perform on for reinforcement learning algorithms. Solutions typically employ algorithms that incentivise environment exploration in different ways.

For the state-of-the art tables, please consult the parent Atari Games task.

( Image credit: Q-map )

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 1–50 of 61 papers

TitleStatusHype
Rainbow: Combining Improvements in Deep Reinforcement LearningCode3
A Study of Global and Episodic Bonuses for Exploration in Contextual MDPsCode1
PoE-World: Compositional World Modeling with Products of Programmatic ExpertsCode1
Cell-Free Latent Go-ExploreCode1
Exploration by Random Network DistillationCode1
First return, then exploreCode1
Flipping Coins to Estimate Pseudocounts for Exploration in Reinforcement LearningCode1
Go-Explore: a New Approach for Hard-Exploration ProblemsCode1
Hybrid RL: Using Both Offline and Online Data Can Make RL EfficientCode1
NovelD: A Simple yet Effective Exploration CriterionCode1
Open-Ended Reinforcement Learning with Neural Reward FunctionsCode1
Playing hard exploration games by watching YouTubeCode1
Redeeming Intrinsic Rewards via Constrained OptimizationCode1
Reinforcement Learning with Latent FlowCode1
Feature Control as Intrinsic Motivation for Hierarchical Reinforcement LearningCode0
Fine-tuning Reinforcement Learning Models is Secretly a Forgetting Mitigation ProblemCode0
Uncertainty-sensitive Learning and Planning with EnsemblesCode0
Learning Abstract Models for Strategic Exploration and Fast Reward TransferCode0
Scaling All-Goals Updates in Reinforcement Learning Using Convolutional Neural NetworksCode0
Playing Atari Games with Deep Reinforcement Learning and Human Checkpoint ReplayCode0
A Study of Plasticity Loss in On-Policy Deep Reinforcement LearningCode0
Beating Atari with Natural Language Guided Reinforcement LearningCode0
Unifying Count-Based Exploration and Intrinsic MotivationCode0
DeepSynth: Automata Synthesis for Automatic Task Segmentation in Deep Reinforcement LearningCode0
Empowerment-driven Exploration using Mutual Information EstimationCode0
Combining Experience Replay with Exploration by Random Network DistillationCode0
Using Natural Language for Reward Shaping in Reinforcement LearningCode0
Count-Based Exploration with Neural Density ModelsCode0
Hierarchical Deep Reinforcement Learning: Integrating Temporal Abstraction and Intrinsic MotivationCode0
Uncertainty - sensitive learning and planning with ensemblesCode0
Exploring Unknown States with Action BalanceCode0
MIME: Mutual Information Minimisation Exploration—0
Observe and Look Further: Achieving Consistent Performance on Atari—0
On Bonus Based Exploration Methods In The Arcade Learning Environment—0
On Bonus-Based Exploration Methods in the Arcade Learning Environment—0
Parametrically Retargetable Decision-Makers Tend To Seek Power—0
Paused Agent Replay Refresh—0
Benchmarking Bonus-Based Exploration Methods on the Arcade Learning Environment—0
Bridging the Gap: Providing Post-Hoc Symbolic Explanations for Sequential Decision-Making Problems with Inscrutable Representations—0
Understanding and Preventing Capacity Loss in Reinforcement Learning—0
Contingency-Aware Exploration in Reinforcement Learning—0
Creativity of AI: Hierarchical Planning Model Learning for Facilitating Deep Reinforcement Learning—0
Curiosity in Hindsight: Intrinsic Exploration in Stochastic Environments—0
Deep Abstract Q-Networks—0
Learning High-level Representations from Demonstrations—0
Memory Based Trajectory-conditioned Policies for Learning from Sparse Rewards—0
Entropic Desired Dynamics for Intrinsic Control—0
Escape Room: A Configurable Testbed for Hierarchical Reinforcement Learning—0
Exploration by Random Network Distillation—0
Exploration in Feature Space for Reinforcement Learning—0
Show:102550
← PrevPage 1 of 2Next →

No leaderboard results yet.