Model Alignment Search

• URL: http://arxiv.org/abs/2501.06164v7
• Date: Fri, 31 Oct 2025 19:40:07 GMT
• Title: Model Alignment Search
• Authors: Satchel Grant,
• Abstract summary: We introduce a method that bidirectionally transfers neural activity between artificial neural networks and uses their resulting behavior as a measure of functional similarity.<n>We show how the method can be used to transfer the behavior from one frozen Neural Network (NN) to another in a manner similar to model stitching.<n>We present a case study on number-related tasks showing that the method can be used to examine specific subtypes of causal information.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: When can we say that two neural systems perform a task in the same way? What nuances do we miss when we fail to causally probe the representations of the systems, and how do we establish bidirectional causal relationships? In this work, we introduce a method that bidirectionally transfers neural activity between artificial neural networks and uses their resulting behavior as a measure of functional similarity. We first show that the method can be used to transfer the behavior from one frozen Neural Network (NN) to another in a manner similar to model stitching, and we show how the method can differ from correlative similarity measures like Representational Similarity Analysis. Next, we empirically and theoretically show how the method can be equivalent to model stitching when desired, or it can take a form that has a more restrictive focus to shared causal information; in both forms, it reduces the number of required matrices for a comparison of n models to be linear in n. We then present a case study on number-related tasks showing that the method can be used to examine specific subtypes of causal information demonstrating that numbers can be encoded differently in recurrent models depending on the task, and we present another case study showing that MAS can reveal misalignment in fine-tuned DeepSeek-r1-Qwen-1.5B models. Lastly, we augment the loss function with a counterfactual latent (CL) auxiliary objective to improve causal relevance when one of the two networks is causally inaccessible (as is often the case in comparisons with biological networks). We use our results to encourage the use of causal methods in neural similarity analyses and to suggest future explorations of network similarity methodology for model misalignment.

Related papers

• Connecting Neural Models Latent Geometries with Relative Geodesic Representations [21.71782603770616]
  We show that when a latent structure is shared between distinct latent spaces, relative distances between representations can be preserved, up to distortions.<n>We assume that distinct neural models parametrize approximately the same underlying manifold, and introduce a representation based on the pullback metric.<n>We validate our method on model stitching and retrieval tasks, covering autoencoders and vision foundation discriminative models.
  arXiv  Detail & Related papers  (2025-06-02T12:34:55Z)
• An XAI-based Analysis of Shortcut Learning in Neural Networks [2.592470112714595]
  We introduce the neuron spurious score to quantify a neuron's dependence on spurious features.<n>Our results show that spurious features are partially disentangled, but the degree of disentanglement varies across model architectures.<n>Our results lay the groundwork for the development of novel methods to mitigate spurious correlations and make AI models safer to use in practice.
  arXiv  Detail & Related papers  (2025-04-22T07:40:45Z)
• Bilinear Convolution Decomposition for Causal RL Interpretability [0.0]
  Efforts to interpret reinforcement learning (RL) models often rely on high-level techniques such as attribution or probing.<n>This work proposes replacing nonlinearities in convolutional neural networks (ConvNets) with bilinear variants, to produce a class of models for which these limitations can be addressed.<n>We show bilinear model variants perform comparably in model-free reinforcement learning settings, and give a side by side comparison on ProcGen environments.
  arXiv  Detail & Related papers  (2024-12-01T19:32:04Z)
• Evaluating Representational Similarity Measures from the Lens of Functional Correspondence [1.7811840395202345]
  Neuroscience and artificial intelligence (AI) both face the challenge of interpreting high-dimensional neural data. Despite the widespread use of representational comparisons, a critical question remains: which metrics are most suitable for these comparisons?
  arXiv  Detail & Related papers  (2024-11-21T23:53:58Z)
• Measuring similarity between embedding spaces using induced neighborhood graphs [10.056989400384772]
  We propose a metric to evaluate the similarity between paired item representations. Our results show that accuracy in both analogy and zero-shot classification tasks correlates with the embedding similarity.
  arXiv  Detail & Related papers  (2024-11-13T15:22:33Z)
• What Representational Similarity Measures Imply about Decodable Information [6.5879381737929945]
  We show that some neural network similarity measures can be equivalently motivated from a decoding perspective. Measures like CKA and CCA quantify the average alignment between optimal linear readouts across a distribution of decoding tasks. Overall, our work demonstrates a tight link between the geometry of neural representations and the ability to linearly decode information.
  arXiv  Detail & Related papers  (2024-11-12T21:37:10Z)
• Linking in Style: Understanding learned features in deep learning models [0.0]
  Convolutional neural networks (CNNs) learn abstract features to perform object classification. We propose an automatic method to visualize and systematically analyze learned features in CNNs.
  arXiv  Detail & Related papers  (2024-09-25T12:28:48Z)
• Understanding Probe Behaviors through Variational Bounds of Mutual Information [53.520525292756005]
  We provide guidelines for linear probing by constructing a novel mathematical framework leveraging information theory. First, we connect probing with the variational bounds of mutual information (MI) to relax the probe design, equating linear probing with fine-tuning. We show that the intermediate representations can have the biggest MI estimate because of the tradeoff between better separability and decreasing MI.
  arXiv  Detail & Related papers  (2023-12-15T18:38:18Z)
• Interpretable Differencing of Machine Learning Models [20.99877540751412]
  We formalize the problem of model differencing as one of predicting a dissimilarity function of two ML models' outputs. A Joint Surrogate Tree (JST) is composed of two conjoined decision tree surrogates for the two models. A JST provides an intuitive representation of differences and places the changes in the context of the models' decision logic.
  arXiv  Detail & Related papers  (2023-06-10T16:15:55Z)
• All Roads Lead to Rome? Exploring the Invariance of Transformers' Representations [69.3461199976959]
  We propose a model based on invertible neural networks, BERT-INN, to learn the Bijection Hypothesis. We show the advantage of BERT-INN both theoretically and through extensive experiments.
  arXiv  Detail & Related papers  (2023-05-23T22:30:43Z)
• Linear Causal Disentanglement via Interventions [8.444187296409051]
  Causal disentanglement seeks a representation of data involving latent variables that relate to one another via a causal model. We study observed variables that are a linear transformation of a linear latent causal model.
  arXiv  Detail & Related papers  (2022-11-29T18:43:42Z)
• Neural Dependencies Emerging from Learning Massive Categories [94.77992221690742]
  This work presents two astonishing findings on neural networks learned for large-scale image classification. 1) Given a well-trained model, the logits predicted for some category can be directly obtained by linearly combining the predictions of a few other categories. 2) Neural dependencies exist not only within a single model, but even between two independently learned models.
  arXiv  Detail & Related papers  (2022-11-21T09:42:15Z)
• Linear Connectivity Reveals Generalization Strategies [54.947772002394736]
  Some pairs of finetuned models have large barriers of increasing loss on the linear paths between them. We find distinct clusters of models which are linearly connected on the test loss surface, but are disconnected from models outside the cluster. Our work demonstrates how the geometry of the loss surface can guide models towards different functions.
  arXiv  Detail & Related papers  (2022-05-24T23:43:02Z)
• Dynamically-Scaled Deep Canonical Correlation Analysis [77.34726150561087]
  Canonical Correlation Analysis (CCA) is a method for feature extraction of two views by finding maximally correlated linear projections of them. We introduce a novel dynamic scaling method for training an input-dependent canonical correlation model.
  arXiv  Detail & Related papers  (2022-03-23T12:52:49Z)
• CCSL: A Causal Structure Learning Method from Multiple Unknown Environments [32.61349047509467]
  We propose a unified Causal Cluster Structures Learning (named CCSL) method for causal discovery from non-i.i.d. data. This method simultaneously integrates the following two tasks: 1) clustering subjects with the same causal mechanism; 2) learning causal structures from the samples of subjects.
  arXiv  Detail & Related papers  (2021-11-18T12:50:53Z)
• The Causal Neural Connection: Expressiveness, Learnability, and Inference [125.57815987218756]
  An object called structural causal model (SCM) represents a collection of mechanisms and sources of random variation of the system under investigation. In this paper, we show that the causal hierarchy theorem (Thm. 1, Bareinboim et al., 2020) still holds for neural models. We introduce a special type of SCM called a neural causal model (NCM), and formalize a new type of inductive bias to encode structural constraints necessary for performing causal inferences.
  arXiv  Detail & Related papers  (2021-07-02T01:55:18Z)
• Counterfactual Invariance to Spurious Correlations: Why and How to Pass Stress Tests [87.60900567941428]
  A spurious correlation' is the dependence of a model on some aspect of the input data that an analyst thinks shouldn't matter. In machine learning, these have a know-it-when-you-see-it character. We study stress testing using the tools of causal inference.
  arXiv  Detail & Related papers  (2021-05-31T14:39:38Z)
• On the Transferability of Adversarial Attacksagainst Neural Text Classifier [121.6758865857686]
  We investigate the transferability of adversarial examples for text classification models. We propose a genetic algorithm to find an ensemble of models that can induce adversarial examples to fool almost all existing models. We derive word replacement rules that can be used for model diagnostics from these adversarial examples.
  arXiv  Detail & Related papers  (2020-11-17T10:45:05Z)
• Interpretable Multi-dataset Evaluation for Named Entity Recognition [110.64368106131062]
  We present a general methodology for interpretable evaluation for the named entity recognition (NER) task. The proposed evaluation method enables us to interpret the differences in models and datasets, as well as the interplay between them. By making our analysis tool available, we make it easy for future researchers to run similar analyses and drive progress in this area.
  arXiv  Detail & Related papers  (2020-11-13T10:53:27Z)
• Multi-task Supervised Learning via Cross-learning [102.64082402388192]
  We consider a problem known as multi-task learning, consisting of fitting a set of regression functions intended for solving different tasks. In our novel formulation, we couple the parameters of these functions, so that they learn in their task specific domains while staying close to each other. This facilitates cross-fertilization in which data collected across different domains help improving the learning performance at each other task.
  arXiv  Detail & Related papers  (2020-10-24T21:35:57Z)
• Category-Learning with Context-Augmented Autoencoder [63.05016513788047]
  Finding an interpretable non-redundant representation of real-world data is one of the key problems in Machine Learning. We propose a novel method of using data augmentations when training autoencoders. We train a Variational Autoencoder in such a way, that it makes transformation outcome predictable by auxiliary network.
  arXiv  Detail & Related papers  (2020-10-10T14:04:44Z)
• Structural Causal Models Are (Solvable by) Credal Networks [70.45873402967297]
  Causal inferences can be obtained by standard algorithms for the updating of credal nets. This contribution should be regarded as a systematic approach to represent structural causal models by credal networks. Experiments show that approximate algorithms for credal networks can immediately be used to do causal inference in real-size problems.
  arXiv  Detail & Related papers  (2020-08-02T11:19:36Z)
• 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)
• Provably Efficient Neural Estimation of Structural Equation Model: An Adversarial Approach [144.21892195917758]
  We study estimation in a class of generalized Structural equation models (SEMs) We formulate the linear operator equation as a min-max game, where both players are parameterized by neural networks (NNs), and learn the parameters of these neural networks using a gradient descent. For the first time we provide a tractable estimation procedure for SEMs based on NNs with provable convergence and without the need for sample splitting.
  arXiv  Detail & Related papers  (2020-07-02T17:55:47Z)
• Pairwise Supervision Can Provably Elicit a Decision Boundary [84.58020117487898]
  Similarity learning is a problem to elicit useful representations by predicting the relationship between a pair of patterns. We show that similarity learning is capable of solving binary classification by directly eliciting a decision boundary.
  arXiv  Detail & Related papers  (2020-06-11T05:35:16Z)
• Learning What Makes a Difference from Counterfactual Examples and Gradient Supervision [57.14468881854616]
  We propose an auxiliary training objective that improves the generalization capabilities of neural networks. We use pairs of minimally-different examples with different labels, a.k.a counterfactual or contrasting examples, which provide a signal indicative of the underlying causal structure of the task. Models trained with this technique demonstrate improved performance on out-of-distribution test sets.
  arXiv  Detail & Related papers  (2020-04-20T02:47:49Z)
• Similarity of Neural Networks with Gradients [8.804507286438781]
  We propose to leverage both feature vectors and gradient ones into designing the representation of a neural network. We show that the proposed approach provides a state-of-the-art method for computing similarity of neural networks.
  arXiv  Detail & Related papers  (2020-03-25T17:04:10Z)

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.