Explaining the Impact of Training on Vision Models via Activation Clustering

• URL: http://arxiv.org/abs/2411.19700v2
• Date: Wed, 19 Feb 2025 18:21:07 GMT
• Title: Explaining the Impact of Training on Vision Models via Activation Clustering
• Authors: Ahcène Boubekki, Samuel G. Fadel, Sebastian Mair,
• Abstract summary: We propose Neuro-Activated Vision Explanations (NAVE), which extracts the information captured by the encoder by clustering the feature activations of the frozen network to be explained. NAVE does not aim to explain the model's prediction but to answer questions such as which parts of the image are processed similarly or which information is kept in deeper layers.
• Score: 2.8792218859042453
• License:
• Abstract: Recent developments in the field of explainable artificial intelligence (XAI) for vision models investigate the information extracted by their feature encoder. We contribute to this effort and propose Neuro-Activated Vision Explanations (NAVE), which extracts the information captured by the encoder by clustering the feature activations of the frozen network to be explained. The method does not aim to explain the model's prediction but to answer questions such as which parts of the image are processed similarly or which information is kept in deeper layers. Experimentally, we leverage NAVE to show that the training dataset and the level of supervision affect which concepts are captured. In addition, our method reveals the impact of registers on vision transformers (ViT) and the information saturation caused by the watermark Clever Hans effect in the training set.

Related papers

• Enhancing Neural Network Interpretability Through Conductance-Based Information Plane Analysis [0.0]
  The Information Plane is a conceptual framework used to analyze the flow of information in neural networks. This paper introduces a new approach that uses layer conductance, a measure of sensitivity to input features, to enhance the Information Plane analysis.
  arXiv  Detail & Related papers  (2024-08-26T23:10:42Z)
• Learning Object-Centric Representation via Reverse Hierarchy Guidance [73.05170419085796]
  Object-Centric Learning (OCL) seeks to enable Neural Networks to identify individual objects in visual scenes. RHGNet introduces a top-down pathway that works in different ways in the training and inference processes. Our model achieves SOTA performance on several commonly used datasets.
  arXiv  Detail & Related papers  (2024-05-17T07:48:27Z)
• Automatic Discovery of Visual Circuits [66.99553804855931]
  We explore scalable methods for extracting the subgraph of a vision model's computational graph that underlies recognition of a specific visual concept. We find that our approach extracts circuits that causally affect model output, and that editing these circuits can defend large pretrained models from adversarial attacks.
  arXiv  Detail & Related papers  (2024-04-22T17:00:57Z)
• Manipulating Feature Visualizations with Gradient Slingshots [54.31109240020007]
  We introduce a novel method for manipulating Feature Visualization (FV) without significantly impacting the model's decision-making process. We evaluate the effectiveness of our method on several neural network models and demonstrate its capabilities to hide the functionality of arbitrarily chosen neurons.
  arXiv  Detail & Related papers  (2024-01-11T18:57:17Z)
• Advancing Ante-Hoc Explainable Models through Generative Adversarial Networks [24.45212348373868]
  This paper presents a novel concept learning framework for enhancing model interpretability and performance in visual classification tasks. Our approach appends an unsupervised explanation generator to the primary classifier network and makes use of adversarial training. This work presents a significant step towards building inherently interpretable deep vision models with task-aligned concept representations.
  arXiv  Detail & Related papers  (2024-01-09T16:16:16Z)
• TVE: Learning Meta-attribution for Transferable Vision Explainer [76.68234965262761]
  We introduce a Transferable Vision Explainer (TVE) that can effectively explain various vision models in downstream tasks. TVE is realized through a pre-training process on large-scale datasets towards learning the meta-attribution. This meta-attribution leverages the versatility of generic backbone encoders to comprehensively encode the attribution knowledge for the input instance, which enables TVE to seamlessly transfer to explain various downstream tasks.
  arXiv  Detail & Related papers  (2023-12-23T21:49:23Z)
• VISIT: Visualizing and Interpreting the Semantic Information Flow of Transformers [45.42482446288144]
  Recent advances in interpretability suggest we can project weights and hidden states of transformer-based language models to their vocabulary. We investigate LM attention heads and memory values, the vectors the models dynamically create and recall while processing a given input. We create a tool to visualize a forward pass of Generative Pre-trained Transformers (GPTs) as an interactive flow graph.
  arXiv  Detail & Related papers  (2023-05-22T19:04:56Z)
• Visualizing Transferred Knowledge: An Interpretive Model of Unsupervised Domain Adaptation [70.85686267987744]
  Unsupervised domain adaptation problems can transfer knowledge from a labeled source domain to an unlabeled target domain. We propose an interpretive model of unsupervised domain adaptation, as the first attempt to visually unveil the mystery of transferred knowledge. Our method provides an intuitive explanation for the base model's predictions and unveils transfer knowledge by matching the image patches with the same semantics across both source and target domains.
  arXiv  Detail & Related papers  (2023-03-04T03:02:12Z)
• A Detailed Study of Interpretability of Deep Neural Network based Top Taggers [3.8541104292281805]
  Recent developments in explainable AI (XAI) allow researchers to explore the inner workings of deep neural networks (DNNs) We explore interpretability of models designed to identify jets coming from top quark decay in high energy proton-proton collisions at the Large Hadron Collider (LHC) Our studies uncover some major pitfalls of existing XAI methods and illustrate how they can be overcome to obtain consistent and meaningful interpretation of these models.
  arXiv  Detail & Related papers  (2022-10-09T23:02:42Z)
• Entity-Conditioned Question Generation for Robust Attention Distribution in Neural Information Retrieval [51.53892300802014]
  We show that supervised neural information retrieval models are prone to learning sparse attention patterns over passage tokens. Using a novel targeted synthetic data generation method, we teach neural IR to attend more uniformly and robustly to all entities in a given passage.
  arXiv  Detail & Related papers  (2022-04-24T22:36:48Z)
• Perception Visualization: Seeing Through the Eyes of a DNN [5.9557391359320375]
  We develop a new form of explanation that is radically different in nature from current explanation methods, such as Grad-CAM. Perception visualization provides a visual representation of what the DNN perceives in the input image by depicting what visual patterns the latent representation corresponds to. Results of our user study demonstrate that humans can better understand and predict the system's decisions when perception visualizations are available.
  arXiv  Detail & Related papers  (2022-04-21T07:18:55Z)

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.