Strong Model Collapse

• URL: http://arxiv.org/abs/2410.04840v2
• Date: Tue, 8 Oct 2024 16:14:43 GMT
• Title: Strong Model Collapse
• Authors: Elvis Dohmatob, Yunzhen Feng, Arjun Subramonian, Julia Kempe,
• Abstract summary: We consider a supervised regression setting and establish the existance of a strong form of the model collapse phenomenon. Our results show that even the smallest fraction of synthetic data can lead to model collapse. We investigate whether increasing model size, an approach aligned with current trends in training large language models, exacerbates or mitigates model collapse.
• Score: 16.071600606637908
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Within the scaling laws paradigm, which underpins the training of large neural networks like ChatGPT and Llama, we consider a supervised regression setting and establish the existance of a strong form of the model collapse phenomenon, a critical performance degradation due to synthetic data in the training corpus. Our results show that even the smallest fraction of synthetic data (e.g., as little as 1\% of the total training dataset) can still lead to model collapse: larger and larger training sets do not enhance performance. We further investigate whether increasing model size, an approach aligned with current trends in training large language models, exacerbates or mitigates model collapse. In a simplified regime where neural networks are approximated via random projections of tunable size, we both theoretically and empirically show that larger models can amplify model collapse. Interestingly, our theory also indicates that, beyond the interpolation threshold (which can be extremely high for very large datasets), larger models may mitigate the collapse, although they do not entirely prevent it. Our theoretical findings are empirically verified through experiments on language models and feed-forward neural networks for images.

Related papers

• Unified Neural Network Scaling Laws and Scale-time Equivalence [10.918504301310753]
  We present a novel theoretical characterization of how three factors -- model size, training time, and data volume -- interact to determine the performance of deep neural networks. We first establish a theoretical and empirical equivalence between scaling the size of a neural network and increasing its training time proportionally. We then combine scale-time equivalence with a linear model analysis of double descent to obtain a unified theoretical scaling law.
  arXiv  Detail & Related papers  (2024-09-09T16:45:26Z)
• Towards Theoretical Understandings of Self-Consuming Generative Models [56.84592466204185]
  This paper tackles the emerging challenge of training generative models within a self-consuming loop. We construct a theoretical framework to rigorously evaluate how this training procedure impacts the data distributions learned by future models. We present results for kernel density estimation, delivering nuanced insights such as the impact of mixed data training on error propagation.
  arXiv  Detail & Related papers  (2024-02-19T02:08:09Z)
• 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)
• Learning to Jump: Thinning and Thickening Latent Counts for Generative Modeling [69.60713300418467]
  Learning to jump is a general recipe for generative modeling of various types of data. We demonstrate when learning to jump is expected to perform comparably to learning to denoise, and when it is expected to perform better.
  arXiv  Detail & Related papers  (2023-05-28T05:38:28Z)
• Training Trajectories of Language Models Across Scales [99.38721327771208]
  Scaling up language models has led to unprecedented performance gains. How do language models of different sizes learn during pre-training? Why do larger language models demonstrate more desirable behaviors?
  arXiv  Detail & Related papers  (2022-12-19T19:16:29Z)
• Closed-form Continuous-Depth Models [99.40335716948101]
  Continuous-depth neural models rely on advanced numerical differential equation solvers. We present a new family of models, termed Closed-form Continuous-depth (CfC) networks, that are simple to describe and at least one order of magnitude faster.
  arXiv  Detail & Related papers  (2021-06-25T22:08:51Z)
• Synthesizing Irreproducibility in Deep Networks [2.28438857884398]
  Modern day deep networks suffer from irreproducibility (also referred to as nondeterminism or underspecification) We show that even with a single nonlinearity and for very simple data and models, irreproducibility occurs. Model complexity and the choice of nonlinearity also play significant roles in making deep models irreproducible.
  arXiv  Detail & Related papers  (2021-02-21T21:51:28Z)
• Exploring Deep Neural Networks via Layer-Peeled Model: Minority Collapse in Imbalanced Training [39.137793683411424]
  We introduce the textitLayer-Peeled Model, a non-yet analytically tractable optimization program. We show that the model inherits many characteristics of well-trained networks, thereby offering an effective tool for explaining and predicting common empirical patterns of deep learning training. In particular, we show that the model reveals a hitherto unknown phenomenon that we term textitMinority Collapse, which fundamentally limits the performance of deep learning models on the minority classes.
  arXiv  Detail & Related papers  (2021-01-29T17:37:17Z)
• Firearm Detection via Convolutional Neural Networks: Comparing a Semantic Segmentation Model Against End-to-End Solutions [68.8204255655161]
  Threat detection of weapons and aggressive behavior from live video can be used for rapid detection and prevention of potentially deadly incidents. One way for achieving this is through the use of artificial intelligence and, in particular, machine learning for image analysis. We compare a traditional monolithic end-to-end deep learning model and a previously proposed model based on an ensemble of simpler neural networks detecting fire-weapons via semantic segmentation.
  arXiv  Detail & Related papers  (2020-12-17T15:19:29Z)
• Forecasting Industrial Aging Processes with Machine Learning Methods [0.0]
  We evaluate a wider range of data-driven models, comparing some traditional stateless models to more complex recurrent neural networks. Our results show that recurrent models produce near perfect predictions when trained on larger datasets.
  arXiv  Detail & Related papers  (2020-02-05T13:06:44Z)

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.