Summary of Efficient Normalized Conformal Prediction and Uncertainty Quantification For Anti-cancer Drug Sensitivity Prediction with Deep Regression Forests, by Daniel Nolte et al.
Efficient Normalized Conformal Prediction and Uncertainty Quantification for Anti-Cancer Drug Sensitivity Prediction with Deep Regression Forests
by Daniel Nolte, Souparno Ghosh, Ranadip Pal
First submitted to arxiv on: 21 Feb 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); 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 Deep learning models are being widely adopted, but they typically provide point predictions without quantifying their uncertainty. This lack of confidence can undermine trust in these critical decision-making tools. Conformal Prediction is a promising approach that pairs machine learning models with prediction intervals, allowing for uncertainty estimation. However, existing methods fail to provide equally accurate intervals for all samples. To address this limitation, we propose a novel method that estimates the uncertainty of each sample by calculating the variance from a Deep Regression Forest. Our results show that this approach improves the efficiency and coverage of normalized inductive conformal prediction on a drug response prediction task. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Deep learning models are really smart computers that make decisions without telling us how sure they are. This can be a problem because we want to know how confident these computers are. One way to solve this is by using something called Conformal Prediction, which gives us an idea of the computer’s uncertainty. The problem is that most methods don’t work well for all cases. To fix this, we came up with a new method that calculates the uncertainty of each sample using a special kind of forest. We tested it on a task involving predicting how drugs affect people and found that our approach works better than existing methods. |
Keywords
* Artificial intelligence * Deep learning * Machine learning * Regression
