Towards Combinatorial Interpretability of Neural Computation

• URL: http://arxiv.org/abs/2504.08842v1
• Date: Thu, 10 Apr 2025 21:28:16 GMT
• Title: Towards Combinatorial Interpretability of Neural Computation
• Authors: Micah Adler, Dan Alistarh, Nir Shavit,
• Abstract summary: We introduce interpretability, a methodology for understanding neural computation by analyzing the computation structures in the sign-based categorization of a network's weights and biases.<n>We demonstrate its power through feature channel coding, a theory that explains how neural networks compute Boolean expressions.
• Score: 36.53010994384343
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: We introduce combinatorial interpretability, a methodology for understanding neural computation by analyzing the combinatorial structures in the sign-based categorization of a network's weights and biases. We demonstrate its power through feature channel coding, a theory that explains how neural networks compute Boolean expressions and potentially underlies other categories of neural network computation. According to this theory, features are computed via feature channels: unique cross-neuron encodings shared among the inputs the feature operates on. Because different feature channels share neurons, the neurons are polysemantic and the channels interfere with one another, making the computation appear inscrutable. We show how to decipher these computations by analyzing a network's feature channel coding, offering complete mechanistic interpretations of several small neural networks that were trained with gradient descent. Crucially, this is achieved via static combinatorial analysis of the weight matrices, without examining activations or training new autoencoding networks. Feature channel coding reframes the superposition hypothesis, shifting the focus from neuron activation directionality in high-dimensional space to the combinatorial structure of codes. It also allows us for the first time to exactly quantify and explain the relationship between a network's parameter size and its computational capacity (i.e. the set of features it can compute with low error), a relationship that is implicitly at the core of many modern scaling laws. Though our initial studies of feature channel coding are restricted to Boolean functions, we believe they provide a rich, controlled, and informative research space, and that the path we propose for combinatorial interpretation of neural computation can provide a basis for understanding both artificial and biological neural circuits.

Related papers

• Coding schemes in neural networks learning classification tasks [52.22978725954347]
  We investigate fully-connected, wide neural networks learning classification tasks. We show that the networks acquire strong, data-dependent features. Surprisingly, the nature of the internal representations depends crucially on the neuronal nonlinearity.
  arXiv  Detail & Related papers  (2024-06-24T14:50:05Z)
• Graph Neural Networks for Learning Equivariant Representations of Neural Networks [55.04145324152541]
  We propose to represent neural networks as computational graphs of parameters. Our approach enables a single model to encode neural computational graphs with diverse architectures. We showcase the effectiveness of our method on a wide range of tasks, including classification and editing of implicit neural representations.
  arXiv  Detail & Related papers  (2024-03-18T18:01:01Z)
• Understanding polysemanticity in neural networks through coding theory [0.8702432681310401]
  We propose a novel practical approach to network interpretability and theoretical insights into polysemanticity and the density of codes. We show how random projections can reveal whether a network exhibits a smooth or non-differentiable code and hence how interpretable the code is. Our approach advances the pursuit of interpretability in neural networks, providing insights into their underlying structure and suggesting new avenues for circuit-level interpretability.
  arXiv  Detail & Related papers  (2024-01-31T16:31:54Z)
• Closed-Form Interpretation of Neural Network Classifiers with Symbolic Gradients [0.7832189413179361]
  I introduce a unified framework for finding a closed-form interpretation of any single neuron in an artificial neural network. I demonstrate how to interpret neural network classifiers to reveal closed-form expressions of the concepts encoded in their decision boundaries.
  arXiv  Detail & Related papers  (2024-01-10T07:47:42Z)
• Permutation Equivariant Neural Functionals [92.0667671999604]
  This work studies the design of neural networks that can process the weights or gradients of other neural networks. We focus on the permutation symmetries that arise in the weights of deep feedforward networks because hidden layer neurons have no inherent order. In our experiments, we find that permutation equivariant neural functionals are effective on a diverse set of tasks.
  arXiv  Detail & Related papers  (2023-02-27T18:52:38Z)
• Efficient, probabilistic analysis of combinatorial neural codes [0.0]
  neural networks encode inputs in the form of combinations of individual neurons' activities. These neural codes present a computational challenge due to their high dimensionality and often large volumes of data. We apply methods previously applied to small examples and apply them to large neural codes generated by experiments.
  arXiv  Detail & Related papers  (2022-10-19T11:58:26Z)
• Data-driven emergence of convolutional structure in neural networks [83.4920717252233]
  We show how fully-connected neural networks solving a discrimination task can learn a convolutional structure directly from their inputs. By carefully designing data models, we show that the emergence of this pattern is triggered by the non-Gaussian, higher-order local structure of the inputs.
  arXiv  Detail & Related papers  (2022-02-01T17:11:13Z)
• Dive into Layers: Neural Network Capacity Bounding using Algebraic Geometry [55.57953219617467]
  We show that the learnability of a neural network is directly related to its size. We use Betti numbers to measure the topological geometric complexity of input data and the neural network. We perform the experiments on a real-world dataset MNIST and the results verify our analysis and conclusion.
  arXiv  Detail & Related papers  (2021-09-03T11:45:51Z)
• Optimal Approximation with Sparse Neural Networks and Applications [0.0]
  We use deep sparsely connected neural networks to measure the complexity of a function class in $L(mathbb Rd)$. We also introduce representation system - a countable collection of functions to guide neural networks. We then analyse the complexity of a class called $beta$ cartoon-like functions using rate-distortion theory and wedgelets construction.
  arXiv  Detail & Related papers  (2021-08-14T05:14:13Z)
• The Connection Between Approximation, Depth Separation and Learnability in Neural Networks [70.55686685872008]
  We study the connection between learnability and approximation capacity. We show that learnability with deep networks of a target function depends on the ability of simpler classes to approximate the target.
  arXiv  Detail & Related papers  (2021-01-31T11:32:30Z)
• A biologically plausible neural network for multi-channel Canonical Correlation Analysis [12.940770779756482]
  Cortical pyramidal neurons receive inputs from multiple neural populations and integrate these inputs in separate dendritic compartments. We seek a multi-channel CCA algorithm that can be implemented in a biologically plausible neural network. For biological plausibility, we require that the network operates in the online setting and its synaptic update rules are local.
  arXiv  Detail & Related papers  (2020-10-01T16:17:53Z)

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.