Summary of Measuring Stochastic Data Complexity with Boltzmann Influence Functions, by Nathan Ng and Roger Grosse and Marzyeh Ghassemi
Measuring Stochastic Data Complexity with Boltzmann Influence Functions
by Nathan Ng, Roger Grosse, Marzyeh Ghassemi
First submitted to arxiv on: 4 Jun 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: None
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 novel method for estimating the uncertainty of a model’s prediction on a test point, using a minimum description length approach. The predictive normalized maximum likelihood (pNML) distribution is used to consider every possible label for a data point and decrease confidence in a prediction if other labels are also consistent with the model and training data. The authors introduce IF-COMP, a scalable and efficient approximation of the pNML distribution that linearizes the model with a temperature-scaled Boltzmann influence function. This approach can be used to produce well-calibrated predictions on test points as well as measure complexity in both labelled and unlabelled settings. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper is about making computers better at guessing what might happen next, even when things get weird. It’s like trying to predict what someone will say next based on what they’ve said before. The new way of doing this uses a special math formula that makes the computer think more carefully about all the possible answers, rather than just picking one. This helps the computer be more accurate and less likely to make mistakes. |
Keywords
» Artificial intelligence » Likelihood » Temperature
