Captum: A unified and generic model interpretability library for PyTorch

• URL: http://arxiv.org/abs/2009.07896v1
• Date: Wed, 16 Sep 2020 18:57:57 GMT
• Title: Captum: A unified and generic model interpretability library for PyTorch
• Authors: Narine Kokhlikyan, Vivek Miglani, Miguel Martin, Edward Wang, Bilal Alsallakh, Jonathan Reynolds, Alexander Melnikov, Natalia Kliushkina, Carlos Araya, Siqi Yan, Orion Reblitz-Richardson
• Abstract summary: We introduce a novel, unified, open-source model interpretability library for PyTorch. The library contains generic implementations of a number of gradient and perturbation-based attribution algorithms. It can be used for both classification and non-classification models.
• Score: 49.72749684393332
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: In this paper we introduce a novel, unified, open-source model interpretability library for PyTorch [12]. The library contains generic implementations of a number of gradient and perturbation-based attribution algorithms, also known as feature, neuron and layer importance algorithms, as well as a set of evaluation metrics for these algorithms. It can be used for both classification and non-classification models including graph-structured models built on Neural Networks (NN). In this paper we give a high-level overview of supported attribution algorithms and show how to perform memory-efficient and scalable computations. We emphasize that the three main characteristics of the library are multimodality, extensibility and ease of use. Multimodality supports different modality of inputs such as image, text, audio or video. Extensibility allows adding new algorithms and features. The library is also designed for easy understanding and use. Besides, we also introduce an interactive visualization tool called Captum Insights that is built on top of Captum library and allows sample-based model debugging and visualization using feature importance metrics.

Related papers

• The Languini Kitchen: Enabling Language Modelling Research at Different Scales of Compute [66.84421705029624]
  We introduce an experimental protocol that enables model comparisons based on equivalent compute, measured in accelerator hours. We pre-process an existing large, diverse, and high-quality dataset of books that surpasses existing academic benchmarks in quality, diversity, and document length. This work also provides two baseline models: a feed-forward model derived from the GPT-2 architecture and a recurrent model in the form of a novel LSTM with ten-fold throughput.
  arXiv  Detail & Related papers  (2023-09-20T10:31:17Z)
• Equivariance with Learned Canonicalization Functions [77.32483958400282]
  We show that learning a small neural network to perform canonicalization is better than using predefineds. Our experiments show that learning the canonicalization function is competitive with existing techniques for learning equivariant functions across many tasks.
  arXiv  Detail & Related papers  (2022-11-11T21:58:15Z)
• TorchNTK: A Library for Calculation of Neural Tangent Kernels of PyTorch Models [16.30276204466139]
  We introduce torchNTK, a python library to calculate the empirical neural tangent kernel (NTK) of neural network models in the PyTorch framework. A feature of the library is that we expose the user to layerwise NTK components, and show that in some regimes a layerwise calculation is more memory efficient.
  arXiv  Detail & Related papers  (2022-05-24T21:27:58Z)
• DomiKnowS: A Library for Integration of Symbolic Domain Knowledge in Deep Learning [12.122347427933637]
  We demonstrate a library for the integration of domain knowledge in deep learning architectures. Using this library, the structure of the data is expressed symbolically via graph declarations. The domain knowledge can be defined explicitly, which improves the models' explainability.
  arXiv  Detail & Related papers  (2021-08-27T16:06:42Z)
• Benchpress: A Scalable and Versatile Workflow for Benchmarking Structure Learning Algorithms [1.7188280334580197]
  Probabilistic graphical models are one common approach to modelling the data generating mechanism. We present a novel Snakemake workflow called Benchpress for producing scalable, reproducible, and platform-independent benchmarks. We demonstrate the applicability of this workflow for learning Bayesian networks in five typical data scenarios.
  arXiv  Detail & Related papers  (2021-07-08T14:19:28Z)
• Efficient Graph Deep Learning in TensorFlow with tf_geometric [53.237754811019464]
  We introduce tf_geometric, an efficient and friendly library for graph deep learning. tf_geometric provides kernel libraries for building Graph Neural Networks (GNNs) as well as implementations of popular GNNs. The kernel libraries consist of infrastructures for building efficient GNNs, including graph data structures, graph map-reduce framework, graph mini-batch strategy, etc.
  arXiv  Detail & Related papers  (2021-01-27T17:16:36Z)
• Towards Efficient Scene Understanding via Squeeze Reasoning [71.1139549949694]
  We propose a novel framework called Squeeze Reasoning. Instead of propagating information on the spatial map, we first learn to squeeze the input feature into a channel-wise global vector. We show that our approach can be modularized as an end-to-end trained block and can be easily plugged into existing networks.
  arXiv  Detail & Related papers  (2020-11-06T12:17:01Z)
• FuncNN: An R Package to Fit Deep Neural Networks Using Generalized Input Spaces [0.0]
  The functional neural network (FuncNN) library is the first such package in any programming language. This paper introduces functions that provide users an avenue to easily build models, generate predictions, and run cross-validations.
  arXiv  Detail & Related papers  (2020-09-18T22:32:29Z)
• An Advance on Variable Elimination with Applications to Tensor-Based Computation [11.358487655918676]
  We present new results on the classical algorithm of variable elimination, which underlies many algorithms including for probabilistic inference. The results relate to exploiting functional dependencies, allowing one to perform inference and learning efficiently on models that have very large treewidth.
  arXiv  Detail & Related papers  (2020-02-21T14:17:44Z)
• Torch-Struct: Deep Structured Prediction Library [138.5262350501951]
  We introduce Torch-Struct, a library for structured prediction. Torch-Struct includes a broad collection of probabilistic structures accessed through a simple and flexible distribution-based API.
  arXiv  Detail & Related papers  (2020-02-03T16:43:02Z)

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.