Summary of The Satml ’24 Cnn Interpretability Competition: New Innovations For Concept-level Interpretability, by Stephen Casper et al.
The SaTML ’24 CNN Interpretability Competition: New Innovations for Concept-Level Interpretability
by Stephen Casper, Jieun Yun, Joonhyuk Baek, Yeseong Jung, Minhwan Kim, Kiwan Kwon, Saerom Park, Hayden Moore, David Shriver, Marissa Connor, Keltin Grimes, Angus Nicolson, Arush Tagade, Jessica Rumbelow, Hieu Minh Nguyen, Dylan Hadfield-Menell
First submitted to arxiv on: 3 Apr 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI)
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 SaTML 2024 CNN Interpretability Competition aimed to develop novel methods for understanding convolutional neural networks (CNNs) at the ImageNet scale, focusing on identifying trojans in CNNs. The competition’s objective was to enable human crowd-workers to reliably diagnose trojans via interpretability tools. Despite the challenge, the featured entries showcased innovative techniques and set a new record on the benchmark from Casper et al., 2023. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The competition aimed to help humans understand AI systems by developing novel methods for interpreting convolutional neural networks (CNNs) at the ImageNet scale. The goal was to enable human crowd-workers to reliably diagnose trojans in CNNs using interpretability tools. Although it remains challenging, the featured entries showcased new techniques and set a record on the benchmark. |
Keywords
* Artificial intelligence * Cnn
