Towards Structured Dynamic Sparse Pre-Training of BERT

• URL: http://arxiv.org/abs/2108.06277v1
• Date: Fri, 13 Aug 2021 14:54:26 GMT
• Title: Towards Structured Dynamic Sparse Pre-Training of BERT
• Authors: Anastasia Dietrich and Frithjof Gressmann and Douglas Orr and Ivan Chelombiev and Daniel Justus and Carlo Luschi
• Abstract summary: We develop and study a straightforward, dynamic always-sparse pre-training approach for BERT language modeling task. We demonstrate that training remains FLOP-efficient when using coarse-grained block sparsity, making it particularly promising for efficient execution on modern hardware accelerators.
• Score: 4.567122178196833
• License: http://creativecommons.org/licenses/by-sa/4.0/
• Abstract: Identifying algorithms for computational efficient unsupervised training of large language models is an important and active area of research. In this work, we develop and study a straightforward, dynamic always-sparse pre-training approach for BERT language modeling task, which leverages periodic compression steps based on magnitude pruning followed by random parameter re-allocation. This approach enables us to achieve Pareto improvements in terms of the number of floating-point operations (FLOPs) over statically sparse and dense models across a broad spectrum of network sizes. Furthermore, we demonstrate that training remains FLOP-efficient when using coarse-grained block sparsity, making it particularly promising for efficient execution on modern hardware accelerators.

Related papers

• Efficient Training of Deep Neural Operator Networks via Randomized Sampling [0.0]
  Deep operator network (DeepNet) has demonstrated success in the real-time prediction of complex dynamics across various scientific and engineering applications. We introduce a random sampling technique to be adopted the training of DeepONet, aimed at improving generalization ability of the model, while significantly reducing computational time. Our results indicate that incorporating randomization in the trunk network inputs during training enhances the efficiency and robustness of DeepONet, offering a promising avenue for improving the framework's performance in modeling complex physical systems.
  arXiv  Detail & Related papers  (2024-09-20T07:18:31Z)
• Adaptive Anomaly Detection in Network Flows with Low-Rank Tensor Decompositions and Deep Unrolling [9.20186865054847]
  Anomaly detection (AD) is increasingly recognized as a key component for ensuring the resilience of future communication systems. This work considers AD in network flows using incomplete measurements. We propose a novel block-successive convex approximation algorithm based on a regularized model-fitting objective. Inspired by Bayesian approaches, we extend the model architecture to perform online adaptation to per-flow and per-time-step statistics.
  arXiv  Detail & Related papers  (2024-09-17T19:59:57Z)
• Building on Efficient Foundations: Effectively Training LLMs with Structured Feedforward Layers [16.253898272659242]
  State-of-the-art results in large language models (LLMs) often rely on scale, which becomes computationally expensive. Our study focuses on transformer-based LLMs, specifically targeting the computationally intensive feedforward networks (FFNs) We show that wide and structured networks can utilize training FLOPs more efficiently, with fewer parameters and lower loss than dense models at their optimal trade-off.
  arXiv  Detail & Related papers  (2024-06-24T08:43:21Z)
• Self-STORM: Deep Unrolled Self-Supervised Learning for Super-Resolution Microscopy [55.2480439325792]
  We introduce deep unrolled self-supervised learning, which alleviates the need for such data by training a sequence-specific, model-based autoencoder. Our proposed method exceeds the performance of its supervised counterparts.
  arXiv  Detail & Related papers  (2024-03-25T17:40:32Z)
• PILOT: A Pre-Trained Model-Based Continual Learning Toolbox [71.63186089279218]
  This paper introduces a pre-trained model-based continual learning toolbox known as PILOT. On the one hand, PILOT implements some state-of-the-art class-incremental learning algorithms based on pre-trained models, such as L2P, DualPrompt, and CODA-Prompt. On the other hand, PILOT fits typical class-incremental learning algorithms within the context of pre-trained models to evaluate their effectiveness.
  arXiv  Detail & Related papers  (2023-09-13T17:55:11Z)
• Intelligence Processing Units Accelerate Neuromorphic Learning [52.952192990802345]
  Spiking neural networks (SNNs) have achieved orders of magnitude improvement in terms of energy consumption and latency. We present an IPU-optimized release of our custom SNN Python package, snnTorch.
  arXiv  Detail & Related papers  (2022-11-19T15:44:08Z)
• Hyper-Learning for Gradient-Based Batch Size Adaptation [2.944323057176686]
  Scheduling the batch size to increase is an effective strategy to control noise when training deep neural networks. We introduce Arbiter as a new hyper-optimization algorithm to perform batch size adaptations for learnable schedulings. We demonstrate Arbiter's effectiveness in several illustrative experiments.
  arXiv  Detail & Related papers  (2022-05-17T11:01:14Z)
• Learning to Continuously Optimize Wireless Resource in a Dynamic Environment: A Bilevel Optimization Perspective [52.497514255040514]
  This work develops a new approach that enables data-driven methods to continuously learn and optimize resource allocation strategies in a dynamic environment. We propose to build the notion of continual learning into wireless system design, so that the learning model can incrementally adapt to the new episodes. Our design is based on a novel bilevel optimization formulation which ensures certain fairness" across different data samples.
  arXiv  Detail & Related papers  (2021-05-03T07:23:39Z)
• Demystifying BERT: Implications for Accelerator Design [4.80595971865854]
  We focus on BERT, one of the most popular NLP transfer learning algorithms, to identify how its algorithmic behavior can guide future accelerator design. We characterize compute-intensive BERT computations and discuss software and possible hardware mechanisms to further optimize these computations. Overall, our analysis identifies holistic solutions to optimize systems for BERT-like models.
  arXiv  Detail & Related papers  (2021-04-14T01:06:49Z)
• Learning to Continuously Optimize Wireless Resource In Episodically Dynamic Environment [55.91291559442884]
  This work develops a methodology that enables data-driven methods to continuously learn and optimize in a dynamic environment. We propose to build the notion of continual learning into the modeling process of learning wireless systems. Our design is based on a novel min-max formulation which ensures certain fairness" across different data samples.
  arXiv  Detail & Related papers  (2020-11-16T08:24:34Z)
• Extrapolation for Large-batch Training in Deep Learning [72.61259487233214]
  We show that a host of variations can be covered in a unified framework that we propose. We prove the convergence of this novel scheme and rigorously evaluate its empirical performance on ResNet, LSTM, and Transformer.
  arXiv  Detail & Related papers  (2020-06-10T08:22:41Z)

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.