Understanding Scaling Laws with Statistical and Approximation Theory for Transformer Neural Networks on Intrinsically Low-dimensional Data

• URL: http://arxiv.org/abs/2411.06646v1
• Date: Mon, 11 Nov 2024 01:05:28 GMT
• Title: Understanding Scaling Laws with Statistical and Approximation Theory for Transformer Neural Networks on Intrinsically Low-dimensional Data
• Authors: Alex Havrilla, Wenjing Liao,
• Abstract summary: In deep neural networks, a model's generalization error is often observed to follow a power scaling law dependent both on the model size and the data size. We show that our theory predicts a power law between the generalization error and both the training data size and the network size for transformers. By leveraging low-dimensional data structures under a manifold hypothesis, we are able to explain transformer scaling laws in a way which respects the data geometry.
• Score: 4.481230230086981
• License:
• Abstract: When training deep neural networks, a model's generalization error is often observed to follow a power scaling law dependent both on the model size and the data size. Perhaps the best known example of such scaling laws are for transformer-based large language models, where networks with billions of parameters are trained on trillions of tokens of text. Yet, despite sustained widespread interest, a rigorous understanding of why transformer scaling laws exist is still missing. To answer this question, we establish novel statistical estimation and mathematical approximation theories for transformers when the input data are concentrated on a low-dimensional manifold. Our theory predicts a power law between the generalization error and both the training data size and the network size for transformers, where the power depends on the intrinsic dimension $d$ of the training data. Notably, the constructed model architecture is shallow, requiring only logarithmic depth in $d$. By leveraging low-dimensional data structures under a manifold hypothesis, we are able to explain transformer scaling laws in a way which respects the data geometry. Moreover, we test our theory with empirical observation by training LLMs on natural language datasets. We find the observed empirical data scaling laws closely agree with our theoretical predictions. Taken together, these results rigorously show the intrinsic dimension of data to be a crucial quantity affecting transformer scaling laws in both theory and practice.

Related papers

• Information-Theoretic Foundations for Neural Scaling Laws [20.617552198581024]
  We develop information-theoretic foundations for neural scaling laws. We observe that the optimal relation between data and model size is linear, up to logarithmic factors.
  arXiv  Detail & Related papers  (2024-06-28T02:20:54Z)
• Scaling Laws in Linear Regression: Compute, Parameters, and Data [86.48154162485712]
  We study the theory of scaling laws in an infinite dimensional linear regression setup. We show that the reducible part of the test error is $Theta(-(a-1) + N-(a-1)/a)$. Our theory is consistent with the empirical neural scaling laws and verified by numerical simulation.
  arXiv  Detail & Related papers  (2024-06-12T17:53:29Z)
• A Tale of Tails: Model Collapse as a Change of Scaling Laws [11.6055501181235]
  We ask: How will the scaling laws change in the inevitable regime where synthetic data makes its way into the training corpus? We develop a theoretical framework of model collapse through the lens of scaling laws. We discover a wide range of decay phenomena, analyzing loss of scaling, shifted scaling with number of generations, the ''un-learning" of skills, and grokking when mixing human and synthesized data.
  arXiv  Detail & Related papers  (2024-02-10T21:06:34Z)
• A Dynamical Model of Neural Scaling Laws [79.59705237659547]
  We analyze a random feature model trained with gradient descent as a solvable model of network training and generalization. Our theory shows how the gap between training and test loss can gradually build up over time due to repeated reuse of data.
  arXiv  Detail & Related papers  (2024-02-02T01:41:38Z)
• Assessing Neural Network Representations During Training Using Noise-Resilient Diffusion Spectral Entropy [55.014926694758195]
  Entropy and mutual information in neural networks provide rich information on the learning process. We leverage data geometry to access the underlying manifold and reliably compute these information-theoretic measures. We show that they form noise-resistant measures of intrinsic dimensionality and relationship strength in high-dimensional simulated data.
  arXiv  Detail & Related papers  (2023-12-04T01:32:42Z)
• DIFFormer: Scalable (Graph) Transformers Induced by Energy Constrained Diffusion [66.21290235237808]
  We introduce an energy constrained diffusion model which encodes a batch of instances from a dataset into evolutionary states. We provide rigorous theory that implies closed-form optimal estimates for the pairwise diffusion strength among arbitrary instance pairs. Experiments highlight the wide applicability of our model as a general-purpose encoder backbone with superior performance in various tasks.
  arXiv  Detail & Related papers  (2023-01-23T15:18:54Z)
• A Solvable Model of Neural Scaling Laws [72.8349503901712]
  Large language models with a huge number of parameters, when trained on near internet-sized number of tokens, have been empirically shown to obey neural scaling laws. We propose a statistical model -- a joint generative data model and random feature model -- that captures this neural scaling phenomenology. Key findings are the manner in which the power laws that occur in the statistics of natural datasets are extended by nonlinear random feature maps.
  arXiv  Detail & Related papers  (2022-10-30T15:13:18Z)
• Data Scaling Laws in NMT: The Effect of Noise and Architecture [59.767899982937756]
  We study the effect of varying the architecture and training data quality on the data scaling properties of Neural Machine Translation (NMT) We find that the data scaling exponents are minimally impacted, suggesting that marginally worse architectures or training data can be compensated for by adding more data.
  arXiv  Detail & Related papers  (2022-02-04T06:53:49Z)
• Explaining Neural Scaling Laws [17.115592382420626]
  Population loss of trained deep neural networks often follows precise power-law scaling relations. We propose a theory that explains the origins of and connects these scaling laws. We identify variance-limited and resolution-limited scaling behavior for both dataset and model size.
  arXiv  Detail & Related papers  (2021-02-12T18:57:46Z)
• Scaling Laws for Transfer [0.5432984841650929]
  We study scaling laws for transfer learning between distributions in an unsupervised, fine-tuning setting. We find that the effective data transferred is described well in the low data regime by a power-law of parameter count and fine-tuning dataset size.
  arXiv  Detail & Related papers  (2021-02-02T04:07:38Z)
• A Neural Scaling Law from the Dimension of the Data Manifold [8.656787568717252]
  When data is plentiful, the loss achieved by well-trained neural networks scales as a power-law $L propto N-alpha$ in the number of network parameters $N$. The scaling law can be explained if neural models are effectively just performing regression on a data manifold of intrinsic dimension $d$. This simple theory predicts that the scaling exponents $alpha approx 4/d$ for cross-entropy and mean-squared error losses.
  arXiv  Detail & Related papers  (2020-04-22T19:16:06Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.