paper-with-me

Papers

Captum: A unified and generic model interpretability library for PyTorch

2020-09-16 · Narine Kokhlikyan, Vivek Miglani, Miguel Martin, Edward Wang, Bilal Alsallakh, Jonathan Reynolds, Alexander Melnikov, Natalia Kliushkina, Carlos Araya, Siqi Yan, Orion Reblitz-Richardson

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.

📄 PDF Abstract BibTeX arXiv:2009.07896

Code (2)

pytorch/captum 공식 구현 pytorch
grahamannett/tellem pytorch

Tasks

Feature Importance

Methods 이 논문이 사용한 방법론

Interpretability 설명 없음

Similar Papers 제목 키워드 기반

Using Captum to Explain Generative Language Models

2023-12-09 · Vivek Miglani, Aobo Yang, Aram H. Markosyan, Diego Garcia-Olano 외

Captum is a comprehensive library for model explainability in PyTorch, offering a range of methods from the interpretability literature to enhance users' understanding of PyTorch models. In this paper, we introduce new f…

Time Interpret: a Unified Model Interpretability Library for Time Series

2023-06-05 · Joseph Enguehard

We introduce $\texttt{time_interpret}$, a library designed as an extension of Captum, with a specific focus on temporal data. As such, this library implements several feature attribution methods that can be used to expla…

Time Series

TDHook: A Lightweight Framework for Interpretability

2025-09-29 · Yoann Poupart arxiv

Interpretability of Deep Neural Networks (DNNs) is a growing field driven by the study of vision and language models. Yet, some use cases, like image captioning, or domains like Deep Reinforcement Learning (DRL), require…

Reinforcement LearningImage Captioning

pyvene: A Library for Understanding and Improving PyTorch Models via Interventions

2024-03-12 · Zhengxuan Wu, Atticus Geiger, Aryaman Arora, Jing Huang 외

Interventions on model-internal states are fundamental operations in many areas of AI, including model editing, steering, robustness, and interpretability. To facilitate such research, we introduce $\textbf{pyvene}$, an …

Model Editing

PyTorch Tabular: A Framework for Deep Learning with Tabular Data

2021-04-28 · Manu Joseph

In spite of showing unreasonable effectiveness in modalities like Text and Image, Deep Learning has always lagged Gradient Boosting in tabular data - both in popularity and performance. But recently there have been newer…

Deep Learning