Summary of Conformal Risk Minimization with Variance Reduction, by Sima Noorani et al.
Conformal Risk Minimization with Variance Reduction
by Sima Noorani, Orlando Romero, Nicolo Dal Fabbro, Hamed Hassani, George J. Pappas
First submitted to arxiv on: 3 Nov 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Machine Learning (stat.ML)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper proposes a new approach to conformal training (CRM) for achieving probabilistic guarantees on black-box models during the training process. The authors identify a source of sample inefficiency in the existing conformal training method, called ConfTr, which leads to noisy estimated gradients and training instability. To address this challenge, they propose variance-reduced conformal training (VR-ConfTr), a CRM method that incorporates a variance reduction technique in the gradient estimation. Experimental results on various benchmark datasets show that VR-ConfTr achieves faster convergence and smaller prediction sets compared to baselines. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Conformal prediction is a way to get accurate guesses from machines without needing extra information about how they work. Normally, this is done after training a model, but some researchers have been working on doing it during training instead. The problem with this approach is that it can be noisy and hard to use in practice. To fix this, the authors created a new way of doing conformal training that reduces noise and makes it more stable. They tested their method on different datasets and showed that it works better than other methods. |
