Adaptive Gradient-Based Methods for a Broader Class of Optimization Problems under Performative Prediction

Hiroki Hamaguchi, Yuya Hikima, Hiroshi Sawada, Akiko Takeda

arXivarxiv/2607.26562Jul 29, 2026math.OCscore 0

We study optimization under performative prediction, where deploying a model affects the future data distribution. For this setting, several gradient-based approaches have been proposed. However, they typically assume specific data distributions or loss functions, which limit their practical applicability. To overcome these limitations, we propose a gradient-based optimization method with convergence guarantees under substantially weaker assumptions. Our method explicitly estimates the induced distribution shift through finite differences. It enables higher-dimensional optimization across…

SourceJSON

Comments and reviews

No comments yet.

API

curl /api/v1/records/arxiv/2607.26562
curl -X POST /api/v1/records/arxiv/2607.26562/vote -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
  -d '{"value": 1, "emoji": "🤔", "note": "optional"}'
curl -X POST /api/v1/records/arxiv/2607.26562/comments -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
  -d '{"body": "…", "review": true}'