On the Surprising Efficacy of Distillation as an Alternative to Pre-Training Small Models

• URL: http://arxiv.org/abs/2404.03263v2
• Date: Fri, 3 May 2024 06:08:30 GMT
• Title: On the Surprising Efficacy of Distillation as an Alternative to Pre-Training Small Models
• Authors: Sean Farhat, Deming Chen,
• Abstract summary: We propose that small models may not need to absorb the cost of pre-training to reap its benefits. We observe that, when distilled on a task from a pre-trained model, a small model can achieve or surpass the performance it would achieve if it was pre-trained then finetuned on that task.
• Score: 7.062887337934677
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: In this paper, we propose that small models may not need to absorb the cost of pre-training to reap its benefits. Instead, they can capitalize on the astonishing results achieved by modern, enormous models to a surprising degree. We observe that, when distilled on a task from a pre-trained teacher model, a small model can achieve or surpass the performance it would achieve if it was pre-trained then finetuned on that task. To allow this phenomenon to be easily leveraged, we establish a connection reducing knowledge distillation to modern contrastive learning, opening two doors: (1) vastly different model architecture pairings can work for the distillation, and (2) most contrastive learning algorithms rooted in the theory of Noise Contrastive Estimation can be easily applied and used. We demonstrate this paradigm using pre-trained teacher models from open-source model hubs, Transformer and convolution based model combinations, and a novel distillation algorithm that massages the Alignment/Uniformity perspective of contrastive learning by Wang & Isola (2020) into a distillation objective. We choose this flavor of contrastive learning due to its low computational cost, an overarching theme of this work. We also observe that this phenomenon tends not to occur if the task is data-limited. However, this can be alleviated by leveraging yet another scale-inspired development: large, pre-trained generative models for dataset augmentation. Again, we use an open-source model, and our rudimentary prompts are sufficient to boost the small model`s performance. Thus, we highlight a training method for small models that is up to 94% faster than the standard pre-training paradigm without sacrificing performance. For practitioners discouraged from fully utilizing modern foundation datasets for their small models due to the prohibitive scale, we believe our work keeps that door open.

Related papers

• Continual Learning of Diffusion Models with Generative Distillation [34.52513912701778]
  Diffusion models are powerful generative models that achieve state-of-the-art performance in image synthesis. In this paper, we propose generative distillation, an approach that distils the entire reverse process of a diffusion model.
  arXiv  Detail & Related papers  (2023-11-23T14:33:03Z)
• Fantastic Gains and Where to Find Them: On the Existence and Prospect of General Knowledge Transfer between Any Pretrained Model [74.62272538148245]
  We show that for arbitrary pairings of pretrained models, one model extracts significant data context unavailable in the other. We investigate if it is possible to transfer such "complementary" knowledge from one model to another without performance degradation.
  arXiv  Detail & Related papers  (2023-10-26T17:59:46Z)
• Reusing Pretrained Models by Multi-linear Operators for Efficient Training [65.64075958382034]
  Training large models from scratch usually costs a substantial amount of resources. Recent studies such as bert2BERT and LiGO have reused small pretrained models to initialize a large model. We propose a method that linearly correlates each weight of the target model to all the weights of the pretrained model.
  arXiv  Detail & Related papers  (2023-10-16T06:16:47Z)
• Gradient Surgery for One-shot Unlearning on Generative Model [0.989293617504294]
  We introduce a simple yet effective approach to remove a data influence on the deep generative model. Inspired by works in multi-task learning, we propose to manipulate gradients to regularize the interplay of influence among samples.
  arXiv  Detail & Related papers  (2023-07-10T13:29:23Z)
• BOOT: Data-free Distillation of Denoising Diffusion Models with Bootstrapping [64.54271680071373]
  Diffusion models have demonstrated excellent potential for generating diverse images. Knowledge distillation has been recently proposed as a remedy that can reduce the number of inference steps to one or a few. We present a novel technique called BOOT, that overcomes limitations with an efficient data-free distillation algorithm.
  arXiv  Detail & Related papers  (2023-06-08T20:30:55Z)
• Towards Efficient Task-Driven Model Reprogramming with Foundation Models [52.411508216448716]
  Vision foundation models exhibit impressive power, benefiting from the extremely large model capacity and broad training data. However, in practice, downstream scenarios may only support a small model due to the limited computational resources or efficiency considerations. This brings a critical challenge for the real-world application of foundation models: one has to transfer the knowledge of a foundation model to the downstream task.
  arXiv  Detail & Related papers  (2023-04-05T07:28:33Z)
• Revealing Secrets From Pre-trained Models [2.0249686991196123]
  Transfer-learning has been widely adopted in many emerging deep learning algorithms. We show that pre-trained models and fine-tuned models have significantly high similarities in weight values. We propose a new model extraction attack that reveals the model architecture and the pre-trained model used by the black-box victim model.
  arXiv  Detail & Related papers  (2022-07-19T20:19:03Z)
• Knowledge distillation: A good teacher is patient and consistent [71.14922743774864]
  There is a growing discrepancy in computer vision between large-scale models that achieve state-of-the-art performance and models that are affordable in practical applications. We identify certain implicit design choices, which may drastically affect the effectiveness of distillation. We obtain a state-of-the-art ResNet-50 model for ImageNet, which achieves 82.8% top-1 accuracy.
  arXiv  Detail & Related papers  (2021-06-09T17:20:40Z)
• Self-Feature Regularization: Self-Feature Distillation Without Teacher Models [0.0]
  Self-Feature Regularization(SFR) is proposed, which uses features in the deep layers to supervise feature learning in the shallow layers. We firstly use generalization-l2 loss to match local features and a many-to-one approach to distill more intensively in the channel dimension.
  arXiv  Detail & Related papers  (2021-03-12T15:29:00Z)
• Model Reuse with Reduced Kernel Mean Embedding Specification [70.044322798187]
  We present a two-phase framework for finding helpful models for a current application. In the upload phase, when a model is uploading into the pool, we construct a reduced kernel mean embedding (RKME) as a specification for the model. Then in the deployment phase, the relatedness of the current task and pre-trained models will be measured based on the value of the RKME specification.
  arXiv  Detail & Related papers  (2020-01-20T15:15:07Z)

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.