Post-Hoc Concept Disentanglement: From Correlated to Isolated Concept Representations

• URL: http://arxiv.org/abs/2503.05522v1
• Date: Fri, 07 Mar 2025 15:45:43 GMT
• Title: Post-Hoc Concept Disentanglement: From Correlated to Isolated Concept Representations
• Authors: Eren Erogullari, Sebastian Lapuschkin, Wojciech Samek, Frederik Pahde,
• Abstract summary: Concept Activation Vectors (CAVs) are widely used to model human-understandable concepts.<n>They are trained by identifying directions from the activations of concept samples to those of non-concept samples.<n>This method produces similar, non-orthogonal directions for correlated concepts, such as "beard" and "necktie"<n>This entanglement complicates the interpretation of concepts in isolation and can lead to undesired effects in CAV applications.
• Score: 12.072112471560716
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Concept Activation Vectors (CAVs) are widely used to model human-understandable concepts as directions within the latent space of neural networks. They are trained by identifying directions from the activations of concept samples to those of non-concept samples. However, this method often produces similar, non-orthogonal directions for correlated concepts, such as "beard" and "necktie" within the CelebA dataset, which frequently co-occur in images of men. This entanglement complicates the interpretation of concepts in isolation and can lead to undesired effects in CAV applications, such as activation steering. To address this issue, we introduce a post-hoc concept disentanglement method that employs a non-orthogonality loss, facilitating the identification of orthogonal concept directions while preserving directional correctness. We evaluate our approach with real-world and controlled correlated concepts in CelebA and a synthetic FunnyBirds dataset with VGG16 and ResNet18 architectures. We further demonstrate the superiority of orthogonalized concept representations in activation steering tasks, allowing (1) the insertion of isolated concepts into input images through generative models and (2) the removal of concepts for effective shortcut suppression with reduced impact on correlated concepts in comparison to baseline CAVs.

Related papers

• OmniPrism: Learning Disentangled Visual Concept for Image Generation [57.21097864811521]
  Creative visual concept generation often draws inspiration from specific concepts in a reference image to produce relevant outcomes. We propose OmniPrism, a visual concept disentangling approach for creative image generation. Our method learns disentangled concept representations guided by natural language and trains a diffusion model to incorporate these concepts.
  arXiv  Detail & Related papers  (2024-12-16T18:59:52Z)
• Improving Intervention Efficacy via Concept Realignment in Concept Bottleneck Models [57.86303579812877]
  Concept Bottleneck Models (CBMs) ground image classification on human-understandable concepts to allow for interpretable model decisions. Existing approaches often require numerous human interventions per image to achieve strong performances. We introduce a trainable concept realignment intervention module, which leverages concept relations to realign concept assignments post-intervention.
  arXiv  Detail & Related papers  (2024-05-02T17:59:01Z)
• Local Concept Embeddings for Analysis of Concept Distributions in DNN Feature Spaces [1.0923877073891446]
  We propose a novel concept analysis framework for deep neural networks (DNNs) Instead of optimizing a single global concept vector on the complete dataset, it generates a local concept embedding (LoCE) vector for each individual sample. Despite its context sensitivity, our method's concept segmentation performance is competitive to global baselines.
  arXiv  Detail & Related papers  (2023-11-24T12:22:00Z)
• Cross-Modal Conceptualization in Bottleneck Models [21.2577097041883]
  Concept Bottleneck Models (CBMs) assume that training examples (e.g., x-ray images) are annotated with high-level concepts. In our approach, we adopt a more moderate assumption and instead use text descriptions, accompanying the images in training, to guide the induction of concepts. Our cross-modal approach treats concepts as discrete latent variables and promotes concepts that (1) are predictive of the label, and (2) can be predicted reliably from both the image and text.
  arXiv  Detail & Related papers  (2023-10-23T11:00:19Z)
• Implicit Concept Removal of Diffusion Models [92.55152501707995]
  Text-to-image (T2I) diffusion models often inadvertently generate unwanted concepts such as watermarks and unsafe images. We present the Geom-Erasing, a novel concept removal method based on the geometric-driven control.
  arXiv  Detail & Related papers  (2023-10-09T17:13:10Z)
• Deep Concept Removal [29.65899467379793]
  We address the problem of concept removal in deep neural networks. We propose a novel method based on adversarial linear classifiers trained on a concept dataset. We also introduce an implicit gradient-based technique to tackle the challenges associated with adversarial training.
  arXiv  Detail & Related papers  (2023-10-09T14:31:03Z)
• Interpretable Neural-Symbolic Concept Reasoning [7.1904050674791185]
  Concept-based models aim to address this issue by learning tasks based on a set of human-understandable concepts. We propose the Deep Concept Reasoner (DCR), the first interpretable concept-based model that builds upon concept embeddings.
  arXiv  Detail & Related papers  (2023-04-27T09:58:15Z)
• Concept Activation Regions: A Generalized Framework For Concept-Based Explanations [95.94432031144716]
  Existing methods assume that the examples illustrating a concept are mapped in a fixed direction of the deep neural network's latent space. In this work, we propose allowing concept examples to be scattered across different clusters in the DNN's latent space. This concept activation region (CAR) formalism yields global concept-based explanations and local concept-based feature importance.
  arXiv  Detail & Related papers  (2022-09-22T17:59:03Z)
• Concept Gradient: Concept-based Interpretation Without Linear Assumption [77.96338722483226]
  Concept Activation Vector (CAV) relies on learning a linear relation between some latent representation of a given model and concepts. We proposed Concept Gradient (CG), extending concept-based interpretation beyond linear concept functions. We demonstrated CG outperforms CAV in both toy examples and real world datasets.
  arXiv  Detail & Related papers  (2022-08-31T17:06:46Z)
• Exploring Concept Contribution Spatially: Hidden Layer Interpretation with Spatial Activation Concept Vector [5.873416857161077]
  Testing with Concept Activation Vector (TCAV) presents a powerful tool to quantify the contribution of query concepts to a target class. For some images where the target object only occupies a small fraction of the region, TCAV evaluation may be interfered with by redundant background features.
  arXiv  Detail & Related papers  (2022-05-21T15:58:57Z)
• Visual Concepts Tokenization [65.61987357146997]
  We propose an unsupervised transformer-based Visual Concepts Tokenization framework, dubbed VCT, to perceive an image into a set of disentangled visual concept tokens. To obtain these concept tokens, we only use cross-attention to extract visual information from the image tokens layer by layer without self-attention between concept tokens. We further propose a Concept Disentangling Loss to facilitate that different concept tokens represent independent visual concepts.
  arXiv  Detail & Related papers  (2022-05-20T11:25:31Z)
• Navigating Neural Space: Revisiting Concept Activation Vectors to Overcome Directional Divergence [14.071950294953005]
  Concept Activation Vectors (CAVs) have emerged as a popular tool for modeling human-understandable concepts in the latent space. In this paper we show that such a separability-oriented leads to solutions, which may diverge from the actual goal of precisely modeling the concept direction. We introduce pattern-based CAVs, solely focussing on concept signals, thereby providing more accurate concept directions.
  arXiv  Detail & Related papers  (2022-02-07T19:40:20Z)

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.