Summary of Lookupffn: Making Transformers Compute-lite For Cpu Inference, by Zhanpeng Zeng et al.
LookupFFN: Making Transformers Compute-lite for CPU inference
by Zhanpeng Zeng, Michael Davies, Pranav Pulijala, Karthikeyan Sankaralingam, Vikas Singh
First submitted to arxiv on: 12 Mar 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 This paper proposes an alternative formulation of GEMM-based Feed Forward Networks (FFNs) that reduces the computational requirements, making it suitable for inference on CPUs. The proposed LookupFFN module leverages the trade-off between compute and memory resources, which is more favorable on CPUs. The authors demonstrate that their approach achieves similar performance to traditional GEMM-based FFNs when pretraining RoBERTa language models, while significantly reducing the floating-point operations required. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper explores a new way to train deep neural networks using CPUs instead of GPUs. It’s like finding an alternative route to get from point A to point B that uses less energy and is more efficient. The authors show that their method can achieve similar results as traditional methods, but it uses much less computational power. This could be important for industries where speed isn’t the top priority, but cost and security are. |
Keywords
* Artificial intelligence * Inference * Pretraining
