mlsanityData to 14 Oct 2026

In a Nutshell, the Human Asked for This: Latent Goals for Following Temporal Specifications

arXiv 2110.09461 ↗First posted 18 Oct 2021cs.AI cs.LGcited by 17

Not on the map. Why?

Abstract (arXiv)

We address the problem of building agents whose goal is to learn to execute out-of distribution (OOD) multi-task instructions expressed in temporal logic (TL) by using deep reinforcement learning (DRL). Recent works provided evidence that the agent's neural architecture is a key feature when DRL agents are learning to solve OOD tasks in TL. Yet, the studies on this topic are still in their infancy. In this work, we propose a new deep learning configuration with inductive biases that lead agents to generate latent representations of their current goal, yielding a stronger generalization performance. We use these latent-goal networks within a neuro-symbolic framework that executes multi-task formally-defined instructions and contrast the performance of the proposed neural networks against employing different state-of-the-art (SOTA) architectures when generalizing to unseen instructions in OOD environments.

Key facts. Cited by 17 papers in the map (data through 14 Oct 2026). Builds on 29 of its 52 references that are in the map. First posted on arXiv 18 Oct 2021.

Builds on

29 of 52 references.

References · 29

  1. 2017Attention Is All You Needcited by 60,361
  2. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  3. 2014Neural Machine Translation by Jointly Learning to Align and Translatecited by 4,548
  4. 2016Asynchronous Methods for Deep Reinforcement Learningcited by 2,258
  5. 2017Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithmcited by 734
  6. 2019The StarCraft Multi-Agent Challengecited by 597
  7. 2017Safe Reinforcement Learning via Shieldingcited by 449
  8. 2019The Neuro-Symbolic Concept Learner: Interpreting Scenes, Words, and Sentences From Natural Supervisioncited by 372
  9. 2017A simple neural network module for relational reasoningcited by 346
  10. 2016Modular Multitask Reinforcement Learning with Policy Sketchescited by 203
  11. 2020The NetHack Learning Environmentcited by 198
  12. 2020One Policy to Control Them All: Shared Modular Policies for Agent-Agnostic Controlcited by 121
17 more
  1. 2019Compositional generalization through meta sequence-to-sequence learningcited by 99
  2. 2017Zero-Shot Task Generalization with Multi-Task Deep Reinforcement Learningcited by 98
  3. 2021LTL2Action: Generalizing LTL Instructions for Multi-Task RLcited by 74
  4. 2020A Composable Specification Language for Reinforcement Learning Taskscited by 69
  5. 2018Relational recurrent neural networkscited by 65
  6. 2019Joint Inference of Reward Machines and Policies for Reinforcement Learningcited by 56
  7. 2020Grounded Language Learning Fast and Slowcited by 54
  8. 2019Environmental drivers of systematicity and generalization in a situated agentcited by 50
  9. 2020Learning to Combine Top-Down and Bottom-Up Signals in Recurrent Neural Networks with Attention over Modulescited by 38
  10. 2020Encoding formulas as deep networks: Reinforcement learning for zero-shot execution of LTL formulascited by 36
  11. 2019An Explicitly Relational Neural Network Architecturecited by 35
  12. 2021Multi-Agent Reinforcement Learning with Temporal Logic Specificationscited by 27
  13. 2021The Logical Options Frameworkcited by 25
  14. 2018Interactive Grounded Language Acquisition and Generalization in a 2D Worldcited by 21
  15. 2020Systematic Generalisation through Task Temporal Logic and Deep Reinforcement Learningcited by 21
  16. 2020Extended Markov Games to Learn Multiple Tasks in Multi-Agent Reinforcement Learningcited by 9
  17. 2020Beyond Tabula-Rasa: a Modular Reinforcement Learning Approach for Physically Embedded 3D Sokobancited by 7

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2022: 3; 2023: 1; 2024: 5; 2025: 4; 2026: 4.

17 papers.

Sort:

By area:

  1. 2024DeepLTL: Learning to Efficiently Satisfy Complex LTL Specifications for Multi-Task RLcited by 21
  2. 2024BlendRL: A Framework for Merging Symbolic and Neural Policy Learningcited by 14
  3. 2025Skill Expansion and Composition in Parameter Spacecited by 14
  4. 2024LTL-Constrained Policy Optimization with Cycle Experience Replaycited by 11
  5. 2024Reward Machines for Deep RL in Noisy and Uncertain Environmentscited by 11
  6. 2022Noisy Symbolic Abstractions for Deep RL: A case study with Reward Machinescited by 10
  7. 2025One Subgoal at a Time: Zero-Shot Generalization to Arbitrary Linear Temporal Logic Requirements in Multi-Task Reinforcement Learningcited by 8
  8. 2022Categorical semantics of compositional reinforcement learningcited by 5
  9. 2023Reinforcement Learning with Knowledge Representation and Reasoning: A Brief Surveycited by 5
  10. 2025Ground-Compose-Reinforce: Grounding Language in Agentic Behaviours using Limited Datacited by 5
  11. 2026PlatoLTL: Scaling LTL-Guided Multi-Task RLcited by 2
  12. 2026Semantically Labelled Automata for Multi-Task Reinforcement Learning with LTL Instructionscited by 2
5 more
  1. 2022Generalizing LTL Instructions via Future Dependent Optionscited by 1
  2. 2024Generalization of Compositional Tasks with Logical Specification via Implicit Planningcited by 1
  3. 2026Zero-Shot Instruction Following in RL via Structured LTL Representationscited by 1
  4. 2025Zero-Shot Instruction Following in RL via Structured LTL Representationscited by 0
  5. 2026Jaxolotl: A Unified High-Performance Benchmark Suite for LTL-Based Multi-Task RLcited by 0