Offline Discovery of Interpretable Skills from Multi-Task Trajectories

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Zhu, Chongyu, Vanniasinghe, Mithun, Chen, Jiayu, Lee, Chi-Guhn
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866915767180591104
author Zhu, Chongyu
Vanniasinghe, Mithun
Chen, Jiayu
Lee, Chi-Guhn
author_facet Zhu, Chongyu
Vanniasinghe, Mithun
Chen, Jiayu
Lee, Chi-Guhn
contents Hierarchical Imitation Learning is a powerful paradigm for acquiring complex robot behaviors from demonstrations. A central challenge, however, lies in discovering reusable skills from long-horizon, multi-task offline data, especially when the data lacks explicit rewards or subtask annotations. In this work, we introduce LOKI, a three-stage end-to-end learning framework designed for offline skill discovery and hierarchical imitation. The framework commences with a two-stage, weakly supervised skill discovery process: Stage one performs coarse, task-aware macro-segmentation by employing an alignment-enforced Vector Quantized VAE guided by weak task labels. Stage two then refines these segments at a micro-level using a self-supervised sequential model, followed by an iterative clustering process to consolidate skill boundaries. The third stage then leverages these precise boundaries to construct a hierarchical policy within an option-based framework-complete with a learned termination condition beta for explicit skill switching. LOKI achieves high success rates on the challenging D4RL Kitchen benchmark and outperforms standard HIL baselines. Furthermore, we demonstrate that the discovered skills are semantically meaningful, aligning with human intuition, and exhibit compositionality by successfully sequencing them to solve a novel, unseen task.
format Preprint
id arxiv_https___arxiv_org_abs_2602_01018
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Offline Discovery of Interpretable Skills from Multi-Task Trajectories
Zhu, Chongyu
Vanniasinghe, Mithun
Chen, Jiayu
Lee, Chi-Guhn
Robotics
Artificial Intelligence
Hierarchical Imitation Learning is a powerful paradigm for acquiring complex robot behaviors from demonstrations. A central challenge, however, lies in discovering reusable skills from long-horizon, multi-task offline data, especially when the data lacks explicit rewards or subtask annotations. In this work, we introduce LOKI, a three-stage end-to-end learning framework designed for offline skill discovery and hierarchical imitation. The framework commences with a two-stage, weakly supervised skill discovery process: Stage one performs coarse, task-aware macro-segmentation by employing an alignment-enforced Vector Quantized VAE guided by weak task labels. Stage two then refines these segments at a micro-level using a self-supervised sequential model, followed by an iterative clustering process to consolidate skill boundaries. The third stage then leverages these precise boundaries to construct a hierarchical policy within an option-based framework-complete with a learned termination condition beta for explicit skill switching. LOKI achieves high success rates on the challenging D4RL Kitchen benchmark and outperforms standard HIL baselines. Furthermore, we demonstrate that the discovered skills are semantically meaningful, aligning with human intuition, and exhibit compositionality by successfully sequencing them to solve a novel, unseen task.
title Offline Discovery of Interpretable Skills from Multi-Task Trajectories
topic Robotics
Artificial Intelligence
url https://arxiv.org/abs/2602.01018