gzip Predicts Data-dependent Scaling Laws
–arXiv.org Artificial Intelligence
Past work has established scaling laws that predict the performance of a neural language model (LM) as a function of its parameter count and the number of tokens it's trained on, enabling optimal allocation of a fixed compute budget. Are these scaling laws agnostic to training data as some prior work suggests? We generate training datasets of varying complexities by modulating the syntactic properties of a PCFG, finding that 1) scaling laws are sensitive to differences in data complexity and that 2) gzip, a compression algorithm, is an effective predictor of how data complexity impacts scaling properties. We propose a new data-dependent scaling law for LM's that accounts for the training data's gzip-compressibility; its compute-optimal frontier increases in dataset size preference (over parameter count preference) as training data becomes harder to compress.
arXiv.org Artificial Intelligence
May-26-2024
- Country:
- North America > United States
- California > San Francisco County
- San Francisco (0.14)
- Louisiana (0.14)
- California > San Francisco County
- North America > United States
- Genre:
- Research Report (1.00)
- Technology: