Back to AI News
arxiv

AgentSSL: Can MLE Agents Leverage Unlabeled Data?

This paper introduces Agentic Semi-Supervised Learning (AgentSSL), which frames semi-supervised learning as a program search problem and tests MLE agents under scarce labeled data and abundant unlabeled data. The agents synthesize code combining established SSL techniques and outperform strong human-engineered baselines in 8 of 12 settings by up to +21.8%. The authors identify program selection as a critical bottleneck and release ProgramSelectBench, a benchmark with 16,977 agent-generated and human-engineered SSL programs. They also evaluate AgentSSL on real-world ecology detection tasks to explore transfer beyond controlled benchmarks.

AI Brief

It presents a concrete new research direction and benchmark for using MLE agents in semi-supervised learning, with measurable gains and a clearly identified bottleneck.

arXiv:2611.17323v1 Announce Type: new Abstract: Applying machine learning systems to new domains introduces bottlenecks in both labeled data acquisition and manual pipeline engineering. While recent MLE agents show promise in automating pipeline design, they have been studied almost exclusively in fully supervised settings, leaving them untested in realistic regimes where labeled data is scarce but unlabeled data is abundant. In this work, we introduce Agentic Semi-Supervised Learning (AgentSSL), framing SSL as a program search problem and studying how agents navigate this space under limited supervision. We find that even under weak feedback and matched compute, agents can synthesize code that combines and adapts established SSL techniques, exceeding the strongest human-engineered baselines in 8 of 12 settings by up to +21.8%. However, our results identify program selection as a critical bottleneck in AgentSSL, where performance drops by as much as 47% when selecting based on the agent's feedback metric. To systematically study this bottleneck, we introduce ProgramSelectBench, a large-scale benchmark with a combined total of 16,977 agent-generated and human-engineered SSL programs, and evaluate both unsupervised and active program selection methods. We further evaluate AgentSSL on real-world ecology detection tasks, highlighting the opportunities and challenges of transferring agent-discovered strategies beyond controlled benchmarks.
Be the first to rate this article.

Comments & Questions (0)

Captcha
Please be respectful — let's keep the conversation friendly.

No comments yet

Be the first to comment!