Saved in:
Bibliographic Details
Main Authors: Wen, Yongyan, Li, Siyuan, Fu, Mingjian, Yang, Yiqin, Wang, Xun, Liu, Peng
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2605.23372
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914591205752832
author Wen, Yongyan
Li, Siyuan
Fu, Mingjian
Yang, Yiqin
Wang, Xun
Liu, Peng
author_facet Wen, Yongyan
Li, Siyuan
Fu, Mingjian
Yang, Yiqin
Wang, Xun
Liu, Peng
contents In curriculum reinforcement learning (CRL), an agent incrementally accumulates knowledge over a sequence of tasks (i.e., a curriculum), and the learning process is aimed at using the accumulated knowledge to finally solve a challenging target task. While early CRL works focus on sequencing candidate tasks, recent research explores automatic curriculum generation. Among the rich CRL literature, the interpolation-based CRL paradigm is a main body, which automatically generates intermediate tasks by interpolating between the initial task distribution and the target task distribution in task space with meaningful distance metrics (i.e., can measure the task similarity). However, in challenging navigation tasks, the non-Euclidean context (task) space invalidates this assumption. To achieve automatic curriculum generation in complex task, we propose a novel automatic curriculum generation approach based on measurable task representation learning. To better measure the similarity, we propose to transform the task space to a latent space. Through a variational autoencoder structure that encodes the reward and the state transitions, we achieve a latent task representation with a task similarity measurement property, and two close task embeddings correspond to two similar tasks in terms of rewards and state transitions. Based on the learned task representation, we further develop an automatic curriculum generation scheme, which can effectively generate new tasks more and more similar to the target task. 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 CRL approaches based on interpolation and generative adversarial networks.
format Preprint
id arxiv_https___arxiv_org_abs_2605_23372
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Curriculum reinforcement learning with measurable task representation learning
Wen, Yongyan
Li, Siyuan
Fu, Mingjian
Yang, Yiqin
Wang, Xun
Liu, Peng
Machine Learning
Artificial Intelligence
In curriculum reinforcement learning (CRL), an agent incrementally accumulates knowledge over a sequence of tasks (i.e., a curriculum), and the learning process is aimed at using the accumulated knowledge to finally solve a challenging target task. While early CRL works focus on sequencing candidate tasks, recent research explores automatic curriculum generation. Among the rich CRL literature, the interpolation-based CRL paradigm is a main body, which automatically generates intermediate tasks by interpolating between the initial task distribution and the target task distribution in task space with meaningful distance metrics (i.e., can measure the task similarity). However, in challenging navigation tasks, the non-Euclidean context (task) space invalidates this assumption. To achieve automatic curriculum generation in complex task, we propose a novel automatic curriculum generation approach based on measurable task representation learning. To better measure the similarity, we propose to transform the task space to a latent space. Through a variational autoencoder structure that encodes the reward and the state transitions, we achieve a latent task representation with a task similarity measurement property, and two close task embeddings correspond to two similar tasks in terms of rewards and state transitions. Based on the learned task representation, we further develop an automatic curriculum generation scheme, which can effectively generate new tasks more and more similar to the target task. 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 CRL approaches based on interpolation and generative adversarial networks.
title Curriculum reinforcement learning with measurable task representation learning
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2605.23372