Summary of Cognitive Evolutionary Learning to Select Feature Interactions For Recommender Systems, by Runlong Yu et al.
Cognitive Evolutionary Learning to Select Feature Interactions for Recommender Systems
by Runlong Yu, Qixiang Shao, Qi Liu, Huan Liu, Enhong Chen
First submitted to arxiv on: 29 May 2024
Categories
- Main: Artificial Intelligence (cs.AI)
- Secondary: Information Retrieval (cs.IR); Neural and Evolutionary Computing (cs.NE)
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 proposed Cognitive EvoLutionary Learning (CELL) framework is a novel approach to adaptively evolve models in commercial recommender systems. The issue with current approaches is that they equally enumerate all features and interactions, which can lead to poor learning abilities of models and unnecessary noise. CELL consists of three stages: DNA search, genome search, and model functioning. It uses the concept of cognitive ability to select appropriate operations, features, and interactions under task guidance. This framework simulates natural selection by diagnosing the fitness of the model on operations, features, and interactions. Experiments on four real-world datasets show that CELL outperforms state-of-the-art baselines. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper proposes a new way to help computers make good recommendations. Right now, computers can be stuck if they have too many features or interactions to consider. This can happen even when experts try to guide the computer’s decision-making process. The authors of this paper want to change that by creating a system that adapts to different tasks and data sets. They call this system CELL, which is like an organism that grows and changes based on its environment. In this case, the environment is made up of features and interactions. The authors tested CELL on real-world data sets and found that it works better than other approaches. |
