CoLoR-Filter: Conditional Loss Reduction Filtering for Targeted Language Model Pre-training

• URL: http://arxiv.org/abs/2406.10670v3
• Date: Tue, 29 Oct 2024 20:26:14 GMT
• Title: CoLoR-Filter: Conditional Loss Reduction Filtering for Targeted Language Model Pre-training
• Authors: David Brandfonbrener, Hanlin Zhang, Andreas Kirsch, Jonathan Richard Schwarz, Sham Kakade,
• Abstract summary: We propose a data selection method, CoLoR-Filter, which leverages an empirical Bayes-inspired approach to derive a simple and computationally efficient selection criterion. CoLoR-Filter can train a 1.2b parameter target model to match a 1.2b parameter model trained on 25b randomly selected tokens with 25x less data for Books and 11x less data for the downstream tasks.
• Score: 10.511388205893295
• License:
• Abstract: Selecting high-quality data for pre-training is crucial in shaping the downstream task performance of language models. A major challenge lies in identifying this optimal subset, a problem generally considered intractable, thus necessitating scalable and effective heuristics. In this work, we propose a data selection method, CoLoR-Filter (Conditional Loss Reduction Filtering), which leverages an empirical Bayes-inspired approach to derive a simple and computationally efficient selection criterion based on the relative loss values of two auxiliary models. In addition to the modeling rationale, we evaluate CoLoR-Filter empirically on two language modeling tasks: (1) selecting data from C4 for domain adaptation to evaluation on Books and (2) selecting data from C4 for a suite of downstream multiple-choice question answering tasks. We demonstrate favorable scaling both as we subselect more aggressively and using small auxiliary models to select data for large target models. As one headline result, CoLoR-Filter data selected using a pair of 150m parameter auxiliary models can train a 1.2b parameter target model to match a 1.2b parameter model trained on 25b randomly selected tokens with 25x less data for Books and 11x less data for the downstream tasks. Code: https://github.com/davidbrandfonbrener/color-filter-olmo Filtered data: https://huggingface.co/datasets/davidbrandfonbrener/color-filtered-c4

Related papers

• ScalingFilter: Assessing Data Quality through Inverse Utilization of Scaling Laws [67.59263833387536]
  ScalingFilter is a novel approach that evaluates text quality based on the perplexity difference between two language models trained on the same data. To assess the bias introduced by quality filtering, we introduce semantic diversity, a metric of utilizing text embedding models for semantic representations.
  arXiv  Detail & Related papers  (2024-08-15T17:59:30Z)
• REFRESH: Responsible and Efficient Feature Reselection Guided by SHAP Values [17.489279048199304]
  REFRESH is a method to reselect features so that additional constraints that are desirable towards model performance can be achieved without having to train several new models. REFRESH's underlying algorithm is a novel technique using SHAP values and correlation analysis that can approximate for the predictions of a model without having to train these models.
  arXiv  Detail & Related papers  (2024-03-13T18:06:43Z)
• DsDm: Model-Aware Dataset Selection with Datamodels [81.01744199870043]
  Standard practice is to filter for examples that match human notions of data quality. We find that selecting according to similarity with "high quality" data sources may not increase (and can even hurt) performance compared to randomly selecting data. Our framework avoids handpicked notions of data quality, and instead models explicitly how the learning process uses train datapoints to predict on the target tasks.
  arXiv  Detail & Related papers  (2024-01-23T17:22:00Z)
• Confidence-Based Model Selection: When to Take Shortcuts for Subpopulation Shifts [119.22672589020394]
  We propose COnfidence-baSed MOdel Selection (CosMoS), where model confidence can effectively guide model selection. We evaluate CosMoS on four datasets with spurious correlations, each with multiple test sets with varying levels of data distribution shift.
  arXiv  Detail & Related papers  (2023-06-19T18:48:15Z)
• Beta-Rank: A Robust Convolutional Filter Pruning Method For Imbalanced Medical Image Analysis [1.3443196224057659]
  Deep neural networks include a high number of parameters and operations. It can be a challenge to implement these models on devices with limited computational resources. We propose a novel filter pruning method by considering the input and output of filters along with the values of the filters that deal with imbalanced datasets better than others.
  arXiv  Detail & Related papers  (2023-04-15T03:05:47Z)
• Data Selection for Language Models via Importance Resampling [90.9263039747723]
  We formalize the problem of selecting a subset of a large raw unlabeled dataset to match a desired target distribution. We extend the classic importance resampling approach used in low-dimensions for LM data selection. We instantiate the DSIR framework with hashed n-gram features for efficiency, enabling the selection of 100M documents in 4.5 hours.
  arXiv  Detail & Related papers  (2023-02-06T23:57:56Z)
• Filter Methods for Feature Selection in Supervised Machine Learning Applications -- Review and Benchmark [0.0]
  This review synthesizes the literature on feature selection benchmarking and evaluates the performance of 58 methods in the widely used R environment. We consider four typical dataset scenarios that are challenging for ML models.
  arXiv  Detail & Related papers  (2021-11-23T20:20:24Z)
• Training Compact CNNs for Image Classification using Dynamic-coded Filter Fusion [139.71852076031962]
  We present a novel filter pruning method, dubbed dynamic-coded filter fusion (DCFF) We derive compact CNNs in a computation-economical and regularization-free manner for efficient image classification. Our DCFF derives a compact VGGNet-16 with only 72.77M FLOPs and 1.06M parameters while reaching top-1 accuracy of 93.47%.
  arXiv  Detail & Related papers  (2021-07-14T18:07:38Z)
• Efficient Data-specific Model Search for Collaborative Filtering [56.60519991956558]
  Collaborative filtering (CF) is a fundamental approach for recommender systems. In this paper, motivated by the recent advances in automated machine learning (AutoML), we propose to design a data-specific CF model. Key here is a new framework that unifies state-of-the-art (SOTA) CF methods and splits them into disjoint stages of input encoding, embedding function, interaction and prediction function.
  arXiv  Detail & Related papers  (2021-06-14T14:30:32Z)

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.