Summary of The Impact Of Quantization and Pruning on Deep Reinforcement Learning Models, by Heng Lu et al.
The Impact of Quantization and Pruning on Deep Reinforcement Learning Models
by Heng Lu, Mehdi Alemi, Reza Rawassizadeh
First submitted to arxiv on: 5 Jul 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 Deep reinforcement learning (DRL) has achieved significant success across various domains, including video games, robotics, and large language models. However, computational costs and memory requirements of DRL models often limit their deployment in resource-constrained environments. Our study investigates the impact of quantization and pruning on DRL models, examining how these techniques influence performance factors such as average return, memory, inference time, and battery utilization across various DRL algorithms and environments. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Deep learning helps computers learn from mistakes. This paper looks at ways to make deep reinforcement learning (DRL) more practical by shrinking the size of the computer programs that do the learning. Two techniques were tested: making numbers smaller (quantization) and removing unnecessary parts (pruning). The results showed that these methods can make DRL models smaller, but they don’t necessarily use less energy or run faster. |
Keywords
» Artificial intelligence » Deep learning » Inference » Pruning » Quantization » Reinforcement learning
