Saved in:
Bibliographic Details
Main Authors: Li, Shihao, Li, Jiachen, Chen, Dongmei
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2604.00388
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912993712799744
author Li, Shihao
Li, Jiachen
Chen, Dongmei
author_facet Li, Shihao
Li, Jiachen
Chen, Dongmei
contents We demonstrate that gradient-based data valuation produces curriculum orderings that significantly outperform metadata-based heuristics for training game-theoretic motion planners. Specifically, we apply TracIn gradient-similarity scoring to GameFormer on the nuPlan benchmark and construct a curriculum that weights training scenarios by their estimated contribution to validation loss reduction. Across three random seeds, the TracIn-weighted curriculum achieves a mean planning ADE of $1.704\pm0.029$\,m, significantly outperforming the metadata-based interaction-difficulty curriculum ($1.822\pm0.014$\,m; paired $t$-test $p=0.021$, Cohen's $d_z=3.88$) while exhibiting lower variance than the uniform baseline ($1.772\pm0.134$\,m). Our analysis reveals that TracIn scores and scenario metadata are nearly orthogonal (Spearman $ρ=-0.014$), indicating that gradient-based valuation captures training dynamics invisible to hand-crafted features. We further show that gradient-based curriculum weighting succeeds where hard data selection fails: TracIn-curated 20\% subsets degrade performance by $2\times$, whereas full-data curriculum weighting with the same scores yields the best results. These findings establish gradient-based data valuation as a practical tool for improving sample efficiency in game-theoretic planning.
format Preprint
id arxiv_https___arxiv_org_abs_2604_00388
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Gradient-Based Data Valuation Improves Curriculum Learning for Game-Theoretic Motion Planning
Li, Shihao
Li, Jiachen
Chen, Dongmei
Machine Learning
Systems and Control
We demonstrate that gradient-based data valuation produces curriculum orderings that significantly outperform metadata-based heuristics for training game-theoretic motion planners. Specifically, we apply TracIn gradient-similarity scoring to GameFormer on the nuPlan benchmark and construct a curriculum that weights training scenarios by their estimated contribution to validation loss reduction. Across three random seeds, the TracIn-weighted curriculum achieves a mean planning ADE of $1.704\pm0.029$\,m, significantly outperforming the metadata-based interaction-difficulty curriculum ($1.822\pm0.014$\,m; paired $t$-test $p=0.021$, Cohen's $d_z=3.88$) while exhibiting lower variance than the uniform baseline ($1.772\pm0.134$\,m). Our analysis reveals that TracIn scores and scenario metadata are nearly orthogonal (Spearman $ρ=-0.014$), indicating that gradient-based valuation captures training dynamics invisible to hand-crafted features. We further show that gradient-based curriculum weighting succeeds where hard data selection fails: TracIn-curated 20\% subsets degrade performance by $2\times$, whereas full-data curriculum weighting with the same scores yields the best results. These findings establish gradient-based data valuation as a practical tool for improving sample efficiency in game-theoretic planning.
title Gradient-Based Data Valuation Improves Curriculum Learning for Game-Theoretic Motion Planning
topic Machine Learning
Systems and Control
url https://arxiv.org/abs/2604.00388