One Backward from Ten Forward, Subsampling for Large-Scale Deep Learning
- URL: http://arxiv.org/abs/2104.13114v1
- Date: Tue, 27 Apr 2021 11:29:02 GMT
- Title: One Backward from Ten Forward, Subsampling for Large-Scale Deep Learning
- Authors: Chaosheng Dong, Xiaojie Jin, Weihao Gao, Yijia Wang, Hongyi Zhang,
Xiang Wu, Jianchao Yang, Xiaobing Liu
- Abstract summary: Large-scale machine learning systems are often continuously trained with enormous data from production environments.
The sheer volume of streaming data poses a significant challenge to real-time training subsystems and ad-hoc sampling is the standard practice.
We propose to record a constant amount of information per instance from these forward passes. The extra information measurably improves the selection of which data instances should participate in forward and backward passes.
- Score: 35.0157090322113
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: Deep learning models in large-scale machine learning systems are often
continuously trained with enormous data from production environments. The sheer
volume of streaming training data poses a significant challenge to real-time
training subsystems and ad-hoc sampling is the standard practice. Our key
insight is that these deployed ML systems continuously perform forward passes
on data instances during inference, but ad-hoc sampling does not take advantage
of this substantial computational effort. Therefore, we propose to record a
constant amount of information per instance from these forward passes. The
extra information measurably improves the selection of which data instances
should participate in forward and backward passes. A novel optimization
framework is proposed to analyze this problem and we provide an efficient
approximation algorithm under the framework of Mini-batch gradient descent as a
practical solution. We also demonstrate the effectiveness of our framework and
algorithm on several large-scale classification and regression tasks, when
compared with competitive baselines widely used in industry.
Related papers
- A Bayesian Approach to Data Point Selection [24.98069363998565]
Data point selection (DPS) is becoming a critical topic in deep learning.
Existing approaches to DPS are predominantly based on a bi-level optimisation (BLO) formulation.
We propose a novel Bayesian approach to DPS.
arXiv Detail & Related papers (2024-11-06T09:04:13Z) - RPS: A Generic Reservoir Patterns Sampler [1.09784964592609]
We introduce an approach that harnesses a weighted reservoir to facilitate direct pattern sampling from streaming batch data.
We present a generic algorithm capable of addressing temporal biases and handling various pattern types, including sequential, weighted, and unweighted itemsets.
arXiv Detail & Related papers (2024-10-31T16:25:21Z) - Adaptive Memory Replay for Continual Learning [29.333341368722653]
Updating Foundation Models as new data becomes available can lead to catastrophic forgetting'
We introduce a framework of adaptive memory replay for continual learning, where sampling of past data is phrased as a multi-armed bandit problem.
We demonstrate the effectiveness of our approach, which maintains high performance while reducing forgetting by up to 10% at no training efficiency cost.
arXiv Detail & Related papers (2024-04-18T22:01:56Z) - Take the Bull by the Horns: Hard Sample-Reweighted Continual Training
Improves LLM Generalization [165.98557106089777]
A key challenge is to enhance the capabilities of large language models (LLMs) amid a looming shortage of high-quality training data.
Our study starts from an empirical strategy for the light continual training of LLMs using their original pre-training data sets.
We then formalize this strategy into a principled framework of Instance-Reweighted Distributionally Robust Optimization.
arXiv Detail & Related papers (2024-02-22T04:10:57Z) - Stochastic Amortization: A Unified Approach to Accelerate Feature and Data Attribution [62.71425232332837]
We show that training amortized models with noisy labels is inexpensive and surprisingly effective.
This approach significantly accelerates several feature attribution and data valuation methods, often yielding an order of magnitude speedup over existing approaches.
arXiv Detail & Related papers (2024-01-29T03:42:37Z) - Enhancing Consistency and Mitigating Bias: A Data Replay Approach for
Incremental Learning [100.7407460674153]
Deep learning systems are prone to catastrophic forgetting when learning from a sequence of tasks.
To mitigate the problem, a line of methods propose to replay the data of experienced tasks when learning new tasks.
However, it is not expected in practice considering the memory constraint or data privacy issue.
As a replacement, data-free data replay methods are proposed by inverting samples from the classification model.
arXiv Detail & Related papers (2024-01-12T12:51:12Z) - Diffusion Generative Flow Samplers: Improving learning signals through
partial trajectory optimization [87.21285093582446]
Diffusion Generative Flow Samplers (DGFS) is a sampling-based framework where the learning process can be tractably broken down into short partial trajectory segments.
Our method takes inspiration from the theory developed for generative flow networks (GFlowNets)
arXiv Detail & Related papers (2023-10-04T09:39:05Z) - AdaSelection: Accelerating Deep Learning Training through Data
Subsampling [27.46630703428186]
We introduce AdaSelection, an adaptive sub-sampling method to identify the most informative sub-samples within each minibatch.
Compared with industry-standard baselines, AdaSelection consistently displays superior performance.
arXiv Detail & Related papers (2023-06-19T07:01:28Z) - CMW-Net: Learning a Class-Aware Sample Weighting Mapping for Robust Deep
Learning [55.733193075728096]
Modern deep neural networks can easily overfit to biased training data containing corrupted labels or class imbalance.
Sample re-weighting methods are popularly used to alleviate this data bias issue.
We propose a meta-model capable of adaptively learning an explicit weighting scheme directly from data.
arXiv Detail & Related papers (2022-02-11T13:49:51Z) - Batch Active Learning at Scale [39.26441165274027]
Batch active learning, which adaptively issues batched queries to a labeling oracle, is a common approach for addressing this problem.
In this work, we analyze an efficient active learning algorithm, which focuses on the large batch setting.
We show that our sampling method, which combines notions of uncertainty and diversity, easily scales to batch sizes (100K-1M) several orders of magnitude larger than used in previous studies.
arXiv Detail & Related papers (2021-07-29T18:14:05Z) - 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.