Search Results for author: Narine Kokhlikyan

Found 13 papers, 3 papers with code

Using Captum to Explain Generative Language Models

no code implementations9 Dec 2023 Vivek Miglani, Aobo Yang, Aram H. Markosyan, Diego Garcia-Olano, Narine Kokhlikyan

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.

Error Discovery by Clustering Influence Embeddings

no code implementations NeurIPS 2023 Fulton Wang, Julius Adebayo, Sarah Tan, Diego Garcia-Olano, Narine Kokhlikyan

We present a method for identifying groups of test examples -- slices -- on which a model under-performs, a task now known as slice discovery.

Clustering

XAIR: A Framework of Explainable AI in Augmented Reality

no code implementations28 Mar 2023 Xuhai Xu, Mengjie Yu, Tanya R. Jonker, Kashyap Todi, Feiyu Lu, Xun Qian, João Marcelo Evangelista Belo, Tianyi Wang, Michelle Li, Aran Mun, Te-Yen Wu, Junxiao Shen, Ting Zhang, Narine Kokhlikyan, Fulton Wang, Paul Sorenson, Sophie Kahyun Kim, Hrvoje Benko

The framework was based on a multi-disciplinary literature review of XAI and HCI research, a large-scale survey probing 500+ end-users' preferences for AR-based explanations, and three workshops with 12 experts collecting their insights about XAI design in AR.

Explainable Artificial Intelligence (XAI)

Bias Mitigation Framework for Intersectional Subgroups in Neural Networks

no code implementations26 Dec 2022 Narine Kokhlikyan, Bilal Alsallakh, Fulton Wang, Vivek Miglani, Oliver Aobo Yang, David Adkins

We propose a fairness-aware learning framework that mitigates intersectional subgroup bias associated with protected attributes.

Fairness

Prescriptive and Descriptive Approaches to Machine-Learning Transparency

no code implementations27 Apr 2022 David Adkins, Bilal Alsallakh, Adeel Cheema, Narine Kokhlikyan, Emily McReynolds, Pushkar Mishra, Chavez Procope, Jeremy Sawruk, Erin Wang, Polina Zvyagina

We further propose a preliminary approach, called Method Cards, which aims to increase the transparency and reproducibility of ML systems by providing prescriptive documentation of commonly-used ML methods and techniques.

BIG-bench Machine Learning Descriptive +2

Convolutional Networks are Inherently Foveated

no code implementations NeurIPS Workshop SVRHM 2021 Bilal Alsallakh, Vivek Miglani, Narine Kokhlikyan, David Adkins, Orion Reblitz-Richardson

When convolutional layers apply no padding, central pixels have more ways to contribute to the convolution than peripheral pixels.

Foveation

Fine-grained Interpretation and Causation Analysis in Deep NLP Models

no code implementations NAACL 2021 Hassan Sajjad, Narine Kokhlikyan, Fahim Dalvi, Nadir Durrani

This paper is a write-up for the tutorial on "Fine-grained Interpretation and Causation Analysis in Deep NLP Models" that we are presenting at NAACL 2021.

Domain Adaptation

Investigating Saturation Effects in Integrated Gradients

1 code implementation23 Oct 2020 Vivek Miglani, Narine Kokhlikyan, Bilal Alsallakh, Miguel Martin, Orion Reblitz-Richardson

We explore these effects and find that gradients in saturated regions of this path, where model output changes minimally, contribute disproportionately to the computed attribution.

Captum: A unified and generic model interpretability library for PyTorch

2 code implementations16 Sep 2020 Narine Kokhlikyan, Vivek Miglani, Miguel Martin, Edward Wang, Bilal Alsallakh, Jonathan Reynolds, Alexander Melnikov, Natalia Kliushkina, Carlos Araya, Siqi Yan, Orion Reblitz-Richardson

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.

Feature Importance

Cannot find the paper you are looking for? You can Submit a new open access paper.