Summary of A Survey Of Lottery Ticket Hypothesis, by Bohan Liu et al.
A Survey of Lottery Ticket Hypothesis
by Bohan Liu, Zijie Zhang, Peixiong He, Zhensen Wang, Yang Xiao, Ruimeng Ye, Yang Zhou, Wei-Shinn Ku, Bo Hui
First submitted to arxiv on: 7 Mar 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: 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 Lottery Ticket Hypothesis (LTH) suggests that dense neural networks contain highly sparse subnetworks, or “winning tickets,” which can outperform the original model when trained alone. Despite empirical and theoretical proofs in many studies, efficiency and scalability remain concerns. The lack of open-source frameworks and a standardized experimental setting hinders further research. This paper surveys previous LTH studies from various perspectives, highlights existing issues, and proposes potential directions for exploration. By examining the state of LTH and establishing a maintained platform for experimentation and benchmarking, this survey aims to advance understanding and development. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The Lottery Ticket Hypothesis says that big neural networks have hidden, very good parts inside them. These “winning tickets” can work even better than the whole network when they’re trained alone. Many people have shown that this idea is true, but there are still some problems to solve. One issue is that it’s hard to make these experiments easy to repeat and compare. This paper looks at all the previous research on LTH in a new way, points out the challenges, and suggests what could be done next. |
