Identifying Sub-networks in Neural Networks via Functionally Similar Representations

• URL: http://arxiv.org/abs/2410.16484v1
• Date: Mon, 21 Oct 2024 20:19:00 GMT
• Title: Identifying Sub-networks in Neural Networks via Functionally Similar Representations
• Authors: Tian Gao, Amit Dhurandhar, Karthikeyan Natesan Ramamurthy, Dennis Wei,
• Abstract summary: We take a step toward automating the understanding of the network by investigating the existence of distinct sub-networks. Our approach offers meaningful insights into the behavior of neural networks with minimal human and computational cost.
• Score: 41.028797971427124
• License:
• Abstract: Mechanistic interpretability aims to provide human-understandable insights into the inner workings of neural network models by examining their internals. Existing approaches typically require significant manual effort and prior knowledge, with strategies tailored to specific tasks. In this work, we take a step toward automating the understanding of the network by investigating the existence of distinct sub-networks. Specifically, we explore a novel automated and task-agnostic approach based on the notion of functionally similar representations within neural networks, reducing the need for human intervention. Our method identifies similar and dissimilar layers in the network, revealing potential sub-components. We achieve this by proposing, for the first time to our knowledge, the use of Gromov-Wasserstein distance, which overcomes challenges posed by varying distributions and dimensionalities across intermediate representations, issues that complicate direct layer-to-layer comparisons. Through experiments on algebraic and language tasks, we observe the emergence of sub-groups within neural network layers corresponding to functional abstractions. Additionally, we find that different training strategies influence the positioning of these sub-groups. Our approach offers meaningful insights into the behavior of neural networks with minimal human and computational cost.

Related papers

• DISCOVER: Making Vision Networks Interpretable via Competition and Dissection [11.028520416752325]
  This work contributes to post-hoc interpretability, and specifically Network Dissection. Our goal is to present a framework that makes it easier to discover the individual functionality of each neuron in a network trained on a vision task.
  arXiv  Detail & Related papers  (2023-10-07T21:57:23Z)
• Understanding Activation Patterns in Artificial Neural Networks by Exploring Stochastic Processes [0.0]
  We propose utilizing the framework of processes, which has been underutilized thus far. We focus solely on activation frequency, leveraging neuroscience techniques used for real neuron spike trains. We derive parameters describing activation patterns in each network, revealing consistent differences across architectures and training sets.
  arXiv  Detail & Related papers  (2023-08-01T22:12:30Z)
• Rank Diminishing in Deep Neural Networks [71.03777954670323]
  Rank of neural networks measures information flowing across layers. It is an instance of a key structural condition that applies across broad domains of machine learning. For neural networks, however, the intrinsic mechanism that yields low-rank structures remains vague and unclear.
  arXiv  Detail & Related papers  (2022-06-13T12:03:32Z)
• Quasi-orthogonality and intrinsic dimensions as measures of learning and generalisation [55.80128181112308]
  We show that dimensionality and quasi-orthogonality of neural networks' feature space may jointly serve as network's performance discriminants. Our findings suggest important relationships between the networks' final performance and properties of their randomly initialised feature spaces.
  arXiv  Detail & Related papers  (2022-03-30T21:47:32Z)
• Interpretable part-whole hierarchies and conceptual-semantic relationships in neural networks [4.153804257347222]
  We present Agglomerator, a framework capable of providing a representation of part-whole hierarchies from visual cues. We evaluate our method on common datasets, such as SmallNORB, MNIST, FashionMNIST, CIFAR-10, and CIFAR-100.
  arXiv  Detail & Related papers  (2022-03-07T10:56:13Z)
• 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)
• A neural anisotropic view of underspecification in deep learning [60.119023683371736]
  We show that the way neural networks handle the underspecification of problems is highly dependent on the data representation. Our results highlight that understanding the architectural inductive bias in deep learning is fundamental to address the fairness, robustness, and generalization of these systems.
  arXiv  Detail & Related papers  (2021-04-29T14:31:09Z)
• Learning Interpretable Models for Coupled Networks Under Domain Constraints [8.308385006727702]
  We investigate the idea of coupled networks by focusing on interactions between structural edges and functional edges of brain networks. We propose a novel formulation to place hard network constraints on the noise term while estimating interactions. We validate our method on multishell diffusion and task-evoked fMRI datasets from the Human Connectome Project.
  arXiv  Detail & Related papers  (2021-04-19T06:23:31Z)
• Understanding the Role of Individual Units in a Deep Neural Network [85.23117441162772]
  We present an analytic framework to systematically identify hidden units within image classification and image generation networks. First, we analyze a convolutional neural network (CNN) trained on scene classification and discover units that match a diverse set of object concepts. Second, we use a similar analytic method to analyze a generative adversarial network (GAN) model trained to generate scenes.
  arXiv  Detail & Related papers  (2020-09-10T17:59:10Z)
• Training spiking neural networks using reinforcement learning [0.0]
  We propose biologically-plausible alternatives to backpropagation to facilitate the training of spiking neural networks. We focus on investigating the candidacy of reinforcement learning rules in solving the spatial and temporal credit assignment problems. We compare and contrast the two approaches by applying them to traditional RL domains such as gridworld, cartpole and mountain car.
  arXiv  Detail & Related papers  (2020-05-12T17:40:36Z)

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.