{
  "id": "2311.11108",
  "version": "v1",
  "published": "2023-11-18T15:50:07.000Z",
  "updated": "2023-11-18T15:50:07.000Z",
  "title": "Auxiliary Losses for Learning Generalizable Concept-based Models",
  "authors": [
    "Ivaxi Sheth",
    "Samira Ebrahimi Kahou"
  ],
  "comment": "Neurips 2023",
  "categories": [
    "cs.LG"
  ],
  "abstract": "The increasing use of neural networks in various applications has lead to increasing apprehensions, underscoring the necessity to understand their operations beyond mere final predictions. As a solution to enhance model transparency, Concept Bottleneck Models (CBMs) have gained popularity since their introduction. CBMs essentially limit the latent space of a model to human-understandable high-level concepts. While beneficial, CBMs have been reported to often learn irrelevant concept representations that consecutively damage model performance. To overcome the performance trade-off, we propose cooperative-Concept Bottleneck Model (coop-CBM). The concept representation of our model is particularly meaningful when fine-grained concept labels are absent. Furthermore, we introduce the concept orthogonal loss (COL) to encourage the separation between the concept representations and to reduce the intra-concept distance. This paper presents extensive experiments on real-world datasets for image classification tasks, namely CUB, AwA2, CelebA and TIL. We also study the performance of coop-CBM models under various distributional shift settings. We show that our proposed method achieves higher accuracy in all distributional shift settings even compared to the black-box models with the highest concept accuracy.",
  "revisions": [
    {
      "version": "v1",
      "updated": "2023-11-18T15:50:07.000Z"
    }
  ],
  "analyses": {
    "keywords": [
      "learning generalizable concept-based models",
      "auxiliary losses",
      "distributional shift settings",
      "method achieves higher accuracy",
      "learn irrelevant concept representations"
    ],
    "note": {
      "typesetting": "TeX",
      "pages": 0,
      "language": "en",
      "license": "arXiv",
      "status": "editable"
    }
  }
}