Exploring Causes of Representational Similarity in Machine Learning Models

• URL: http://arxiv.org/abs/2505.13899v1
• Date: Tue, 20 May 2025 04:03:11 GMT
• Title: Exploring Causes of Representational Similarity in Machine Learning Models
• Authors: Zeyu Michael Li, Hung Anh Vu, Damilola Awofisayo, Emily Wenger,
• Abstract summary: This work explores how two possible causal factors -- dataset overlap and task overlap -- influence downstream model similarity.<n>The exploration of dataset overlap is motivated by the reality that large-scale generative AI models are often trained on overlapping datasets of scraped internet data.<n>We find that both positively correlate with higher representational similarity and that combining them provides the strongest effect.
• Score: 5.976564391252357
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Numerous works have noted significant similarities in how machine learning models represent the world, even across modalities. Although much effort has been devoted to uncovering properties and metrics on which these models align, surprisingly little work has explored causes of this similarity. To advance this line of inquiry, this work explores how two possible causal factors -- dataset overlap and task overlap -- influence downstream model similarity. The exploration of dataset overlap is motivated by the reality that large-scale generative AI models are often trained on overlapping datasets of scraped internet data, while the exploration of task overlap seeks to substantiate claims from a recent work, the Platonic Representation Hypothesis, that task similarity may drive model similarity. We evaluate the effects of both factors through a broad set of experiments. We find that both positively correlate with higher representational similarity and that combining them provides the strongest effect. Our code and dataset are published.

Related papers

• Enhancing Model Fairness and Accuracy with Similarity Networks: A Methodological Approach [0.20718016474717196]
  We use different techniques to map instances into a similarity feature space. Our method's ability to adjust the resolution of pairwise similarity provides clear insights into the relationship between the dataset classification complexity and model fairness.
  arXiv  Detail & Related papers  (2024-11-08T15:43:01Z)
• Training objective drives the consistency of representational similarity across datasets [19.99817888941361]
  The Platonic Representation Hypothesis claims that recent foundation models are converging to a shared representation space as a function of their downstream task performance. Here, we propose a systematic way to measure how representational similarity between models varies with the set of stimuli used to construct the representations. We find that the objective function is the most crucial factor in determining the consistency of representational similarities across datasets.
  arXiv  Detail & Related papers  (2024-11-08T13:35:45Z)
• Learning Defect Prediction from Unrealistic Data [57.53586547895278]
  Pretrained models of code have become popular choices for code understanding and generation tasks. Such models tend to be large and require commensurate volumes of training data. It has become popular to train models with far larger but less realistic datasets, such as functions with artificially injected bugs. Models trained on such data tend to only perform well on similar data, while underperforming on real world programs.
  arXiv  Detail & Related papers  (2023-11-02T01:51:43Z)
• Stubborn Lexical Bias in Data and Models [50.79738900885665]
  We use a new statistical method to examine whether spurious patterns in data appear in models trained on the data. We apply an optimization approach to *reweight* the training data, reducing thousands of spurious correlations. Surprisingly, though this method can successfully reduce lexical biases in the training data, we still find strong evidence of corresponding bias in the trained models.
  arXiv  Detail & Related papers  (2023-06-03T20:12:27Z)
• SALUDA: Surface-based Automotive Lidar Unsupervised Domain Adaptation [62.889835139583965]
  We introduce an unsupervised auxiliary task of learning an implicit underlying surface representation simultaneously on source and target data. As both domains share the same latent representation, the model is forced to accommodate discrepancies between the two sources of data. Our experiments demonstrate that our method achieves a better performance than the current state of the art, both in real-to-real and synthetic-to-real scenarios.
  arXiv  Detail & Related papers  (2023-04-06T17:36:23Z)
• Counterfactual Generation Under Confounding [24.503075567519048]
  A machine learning model, under the influence of observed or unobserved confounders in the training data, can learn spurious correlations. We propose a counterfactual generation method that learns to modify the value of any attribute in an image and generate new images given a set of observed attributes. Our method is computationally efficient, simple to implement, and works well for any number of generative factors and confounding variables.
  arXiv  Detail & Related papers  (2022-10-22T06:39:22Z)
• Investigate the Essence of Long-Tailed Recognition from a Unified Perspective [11.080317683184363]
  deep recognition models often suffer from long-tailed data distributions due to heavy imbalanced sample number across categories. In this work, we demonstrate that long-tailed recognition suffers from both sample number and category similarity.
  arXiv  Detail & Related papers  (2021-07-08T11:08:40Z)
• On the Efficacy of Adversarial Data Collection for Question Answering: Results from a Large-Scale Randomized Study [65.17429512679695]
  In adversarial data collection (ADC), a human workforce interacts with a model in real time, attempting to produce examples that elicit incorrect predictions. Despite ADC's intuitive appeal, it remains unclear when training on adversarial datasets produces more robust models.
  arXiv  Detail & Related papers  (2021-06-02T00:48:33Z)
• Few-shot Visual Reasoning with Meta-analogical Contrastive Learning [141.2562447971]
  We propose to solve a few-shot (or low-shot) visual reasoning problem, by resorting to analogical reasoning. We extract structural relationships between elements in both domains, and enforce them to be as similar as possible with analogical learning. We validate our method on RAVEN dataset, on which it outperforms state-of-the-art method, with larger gains when the training data is scarce.
  arXiv  Detail & Related papers  (2020-07-23T14:00:34Z)
• Amortized Causal Discovery: Learning to Infer Causal Graphs from Time-Series Data [63.15776078733762]
  We propose Amortized Causal Discovery, a novel framework to learn to infer causal relations from time-series data. We demonstrate experimentally that this approach, implemented as a variational model, leads to significant improvements in causal discovery performance.
  arXiv  Detail & Related papers  (2020-06-18T19:59:12Z)
• Building and Interpreting Deep Similarity Models [0.0]
  We propose to make similarities interpretable by augmenting them with an explanation in terms of input features. We develop BiLRP, a scalable and theoretically founded method to systematically decompose similarity scores on pairs of input features.
  arXiv  Detail & Related papers  (2020-03-11T17:46:55Z)

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.