Related papers: Training Neural Networks for Modularity aids Interpretability

Training Neural Networks for Modularity aids Interpretability

URL: http://arxiv.org/abs/2409.15747v1
Date: Tue, 24 Sep 2024 05:03:49 GMT
Title: Training Neural Networks for Modularity aids Interpretability
Authors: Satvik Golechha, Dylan Cope, Nandi Schoots,
Abstract summary: An approach to improve network interpretability is via clusterability, i.e., splitting a model into disjoint clusters that can be studied independently. We find pretrained models to be highly unclusterable and thus train models to be more modular using an enmeshment loss'' function that encourages the formation of non-interacting clusters.
Score: 0.6749750044497732
License: http://creativecommons.org/licenses/by-nc-sa/4.0/
Abstract: An approach to improve network interpretability is via clusterability, i.e., splitting a model into disjoint clusters that can be studied independently. We find pretrained models to be highly unclusterable and thus train models to be more modular using an ``enmeshment loss'' function that encourages the formation of non-interacting clusters. Using automated interpretability measures, we show that our method finds clusters that learn different, disjoint, and smaller circuits for CIFAR-10 labels. Our approach provides a promising direction for making neural networks easier to interpret.

Related papers

Modular Training of Neural Networks aids Interpretability [45.8172254436063]
We define a measure for clusterability and show that pre-trained models form highly enmeshed clusters via spectral graph clustering. Using automated interpretability techniques, we show that our method can help train models that are more modular and learn different, disjoint, and smaller circuits.
arXiv Detail & Related papers (2025-02-04T16:44:38Z)
Harnessing Neural Unit Dynamics for Effective and Scalable Class-Incremental Learning [38.09011520275557]
Class-incremental learning (CIL) aims to train a model to learn new classes from non-stationary data streams without forgetting old ones. We propose a new kind of connectionist model by tailoring neural unit dynamics that adapt the behavior of neural networks for CIL.
arXiv Detail & Related papers (2024-06-04T15:47:03Z)
Is it all a cluster game? -- Exploring Out-of-Distribution Detection based on Clustering in the Embedding Space [7.856998585396422]
It is essential for safety-critical applications of deep neural networks to determine when new inputs are significantly different from the training distribution. We study the structure and separation of clusters in the embedding space and find that supervised contrastive learning leads to well-separated clusters. In our analysis of different training methods, clustering strategies, distance metrics, and thresholding approaches, we observe that there is no clear winner.
arXiv Detail & Related papers (2022-03-16T11:22:23Z)
Self-Ensembling GAN for Cross-Domain Semantic Segmentation [107.27377745720243]
This paper proposes a self-ensembling generative adversarial network (SE-GAN) exploiting cross-domain data for semantic segmentation. In SE-GAN, a teacher network and a student network constitute a self-ensembling model for generating semantic segmentation maps, which together with a discriminator, forms a GAN. Despite its simplicity, we find SE-GAN can significantly boost the performance of adversarial training and enhance the stability of the model.
arXiv Detail & Related papers (2021-12-15T09:50:25Z)
Deep Attention-guided Graph Clustering with Dual Self-supervision [49.040136530379094]
We propose a novel method, namely deep attention-guided graph clustering with dual self-supervision (DAGC) We develop a dual self-supervision solution consisting of a soft self-supervision strategy with a triplet Kullback-Leibler divergence loss and a hard self-supervision strategy with a pseudo supervision loss. Our method consistently outperforms state-of-the-art methods on six benchmark datasets.
arXiv Detail & Related papers (2021-11-10T06:53:03Z)
Meta-learning representations for clustering with infinite Gaussian mixture models [39.56814839510978]
We propose a meta-learning method that train neural networks for obtaining representations such that clustering performance improves. The proposed method can cluster unseen unlabeled data using knowledge meta-learned with labeled data that are different from the unlabeled data.
arXiv Detail & Related papers (2021-03-01T02:05:31Z)
Local Critic Training for Model-Parallel Learning of Deep Neural Networks [94.69202357137452]
We propose a novel model-parallel learning method, called local critic training. We show that the proposed approach successfully decouples the update process of the layer groups for both convolutional neural networks (CNNs) and recurrent neural networks (RNNs) We also show that trained networks by the proposed method can be used for structural optimization.
arXiv Detail & Related papers (2021-02-03T09:30:45Z)
Distributed Training of Deep Learning Models: A Taxonomic Perspective [11.924058430461216]
Distributed deep learning systems (DDLS) train deep neural network models by utilizing the distributed resources of a cluster. We aim to shine some light on the fundamental principles that are at work when training deep neural networks in a cluster of independent machines.
arXiv Detail & Related papers (2020-07-08T08:56:58Z)
Rethinking Clustering for Robustness [56.14672993686335]
ClusTR is a clustering-based and adversary-free training framework to learn robust models. textitClusTR outperforms adversarially-trained networks by up to $4%$ under strong PGD attacks.
arXiv Detail & Related papers (2020-06-13T16:55:51Z)
Model Fusion via Optimal Transport [64.13185244219353]
We present a layer-wise model fusion algorithm for neural networks. We show that this can successfully yield "one-shot" knowledge transfer between neural networks trained on heterogeneous non-i.i.d. data.
arXiv Detail & Related papers (2019-10-12T22:07:15Z)

This list is automatically generated from the titles and abstracts of the papers in this site.