Beyond Fixed Tasks: Unsupervised Environment Design for Task-Level Pairs

• URL: http://arxiv.org/abs/2511.12706v1
• Date: Sun, 16 Nov 2025 17:48:09 GMT
• Title: Beyond Fixed Tasks: Unsupervised Environment Design for Task-Level Pairs
• Abstract summary: ATLAS (Aligning Tasks and Levels for Autocurricula of Specifications) is a novel method that generates joint autocurricula over tasks and levels.<n>Our approach builds upon UED to automatically produce solvable yet challenging task-level pairs for policy training.
• Score: 40.020192680979854
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Training general agents to follow complex instructions (tasks) in intricate environments (levels) remains a core challenge in reinforcement learning. Random sampling of task-level pairs often produces unsolvable combinations, highlighting the need to co-design tasks and levels. While unsupervised environment design (UED) has proven effective at automatically designing level curricula, prior work has only considered a fixed task. We present ATLAS (Aligning Tasks and Levels for Autocurricula of Specifications), a novel method that generates joint autocurricula over tasks and levels. Our approach builds upon UED to automatically produce solvable yet challenging task-level pairs for policy training. To evaluate ATLAS and drive progress in the field, we introduce an evaluation suite that models tasks as reward machines in Minigrid levels. Experiments demonstrate that ATLAS vastly outperforms random sampling approaches, particularly when sampling solvable pairs is unlikely. We further show that mutations leveraging the structure of both tasks and levels accelerate convergence to performant policies.

Related papers

• Curriculum reinforcement learning with measurable task representation learning [17.66199732810559]
  In curriculum reinforcement learning (CRL), an agent accumulates knowledge over a sequence of tasks (i.e., a curriculum)<n>We propose a novel automatic curriculum generation approach based on measurable task representation learning.<n>We evaluate our method in a variety of challenging navigation tasks, and the experiment results indicate that the proposed approach surpasses state-of-the-art generative adversarial approaches.
  arXiv  Detail & Related papers  (2026-05-22T08:36:49Z)
• CoordField: Coordination Field for Agentic UAV Task Allocation In Low-altitude Urban Scenarios [18.95579655633218]
  CoordField is a coordination field agent system for coordinating heterogeneous drone swarms in complex urban scenarios.<n>LLMs is responsible for interpreting high-level human instructions and converting them into executable commands for the UAV swarms.<n>A Coordination field mechanism is proposed to guide UAV motion and task selection, enabling decentralized and adaptive allocation of emergent tasks.
  arXiv  Detail & Related papers  (2025-04-30T18:02:45Z)
• Logical Specifications-guided Dynamic Task Sampling for Reinforcement Learning Agents [9.529492371336286]
  Reinforcement Learning (RL) has made significant strides in enabling artificial agents to learn diverse behaviors. We propose a novel approach, called Logical Specifications-guided Dynamic Task Sampling (LSTS) LSTS learns a set of RL policies to guide an agent from an initial state to a goal state based on a high-level task specification.
  arXiv  Detail & Related papers  (2024-02-06T04:00:21Z)
• Automaton-Guided Curriculum Generation for Reinforcement Learning Agents [14.20447398253189]
  Automaton-guided Curriculum Learning (AGCL) is a novel method for automatically generating curricula for the target task in the form of Directed Acyclic Graphs (DAGs) AGCL encodes the specification in the form of a deterministic finite automaton (DFA), and then uses the DFA along with the Object-Oriented MDP representation to generate a curriculum as a DAG. Experiments in gridworld and physics-based simulated robotics domains show that the curricula produced by AGCL achieve improved time-to-threshold performance.
  arXiv  Detail & Related papers  (2023-04-11T15:14:31Z)
• Reinforcement Learning with Success Induced Task Prioritization [68.8204255655161]
  We introduce Success Induced Task Prioritization (SITP), a framework for automatic curriculum learning. The algorithm selects the order of tasks that provide the fastest learning for agents. We demonstrate that SITP matches or surpasses the results of other curriculum design methods.
  arXiv  Detail & Related papers  (2022-12-30T12:32:43Z)
• CLUTR: Curriculum Learning via Unsupervised Task Representation Learning [130.79246770546413]
  CLUTR is a novel curriculum learning algorithm that decouples task representation and curriculum learning into a two-stage optimization. We show CLUTR outperforms PAIRED, a principled and popular UED method, in terms of generalization and sample efficiency in the challenging CarRacing and navigation environments.
  arXiv  Detail & Related papers  (2022-10-19T01:45:29Z)
• Environment Generation for Zero-Shot Compositional Reinforcement Learning [105.35258025210862]
  Compositional Design of Environments (CoDE) trains a Generator agent to automatically build a series of compositional tasks tailored to the agent's current skill level. We learn to generate environments composed of multiple pages or rooms, and train RL agents capable of completing wide-range of complex tasks in those environments. CoDE yields 4x higher success rate than the strongest baseline, and demonstrates strong performance of real websites learned on 3500 primitive tasks.
  arXiv  Detail & Related papers  (2022-01-21T21:35:01Z)
• Multi-Task Learning with Sequence-Conditioned Transporter Networks [67.57293592529517]
  We aim to solve multi-task learning through the lens of sequence-conditioning and weighted sampling. We propose a new suite of benchmark aimed at compositional tasks, MultiRavens, which allows defining custom task combinations. Second, we propose a vision-based end-to-end system architecture, Sequence-Conditioned Transporter Networks, which augments Goal-Conditioned Transporter Networks with sequence-conditioning and weighted sampling.
  arXiv  Detail & Related papers  (2021-09-15T21:19:11Z)
• Adaptive Procedural Task Generation for Hard-Exploration Problems [78.20918366839399]
  We introduce Adaptive Procedural Task Generation (APT-Gen) to facilitate reinforcement learning in hard-exploration problems. At the heart of our approach is a task generator that learns to create tasks from a parameterized task space via a black-box procedural generation module. To enable curriculum learning in the absence of a direct indicator of learning progress, we propose to train the task generator by balancing the agent's performance in the generated tasks and the similarity to the target tasks.
  arXiv  Detail & Related papers  (2020-07-01T09:38:51Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.