Search Results for author: Karthikeyan Natesan Ramamurthy

Found 48 papers, 9 papers with code

Higher-Order Attention Networks

no code implementations1 Jun 2022 Mustafa Hajij, Ghada Zamzmi, Theodore Papamarkou, Nina Miolane, Aldo Guzmán-Sáenz, Karthikeyan Natesan Ramamurthy

This paper introduces higher-order attention networks (HOANs), a novel class of attention-based neural networks defined on a generalized higher-order domain called a combinatorial complex (CC).

Graph Learning

Analogies and Feature Attributions for Model Agnostic Explanation of Similarity Learners

no code implementations2 Feb 2022 Karthikeyan Natesan Ramamurthy, Amit Dhurandhar, Dennis Wei, Zaid Bin Tariq

We first propose a method that provides feature attributions to explain the similarity between a pair of inputs as determined by a black box similarity learner.

A label efficient two-sample test

1 code implementation17 Nov 2021 Weizhi Li, Gautam Dasarathy, Karthikeyan Natesan Ramamurthy, Visar Berisha

Two-sample tests evaluate whether two samples are realizations of the same distribution (the null hypothesis) or two different distributions (the alternative hypothesis).

Two-sample testing

Data-Centric AI Requires Rethinking Data Notion

no code implementations6 Oct 2021 Mustafa Hajij, Ghada Zamzmi, Karthikeyan Natesan Ramamurthy, Aldo Guzman Saenz

The transition towards data-centric AI requires revisiting data notions from mathematical and implementational standpoints to obtain unified data-centric machine learning packages.

Locally Invariant Explanations: Towards Causal Explanations through Local Invariant Learning

no code implementations29 Sep 2021 Amit Dhurandhar, Karthikeyan Natesan Ramamurthy, Kartik Ahuja, Vijay Arya

Locally interpretable model agnostic explanations (LIME) method is one of the most popular methods used to explain black-box models at a per example level.

Out-of-Distribution Generalization

Your fairness may vary: Pretrained language model fairness in toxic text classification

no code implementations Findings (ACL) 2022 Ioana Baldini, Dennis Wei, Karthikeyan Natesan Ramamurthy, Mikhail Yurochkin, Moninder Singh

Through the analysis of more than a dozen pretrained language models of varying sizes on two toxic text classification tasks (English), we demonstrate that focusing on accuracy measures alone can lead to models with wide variation in fairness characteristics.

Fairness Language Modelling +3

Augmenting Molecular Deep Generative Models with Topological Data Analysis Representations

no code implementations8 Jun 2021 Yair Schiff, Vijil Chenthamarakshan, Samuel Hoffman, Karthikeyan Natesan Ramamurthy, Payel Das

Deep generative models have emerged as a powerful tool for learning useful molecular representations and designing novel molecules with desired properties, with applications in drug discovery and material design.

Drug Discovery Topological Data Analysis

Uncertainty Quantification 360: A Holistic Toolkit for Quantifying and Communicating the Uncertainty of AI

1 code implementation2 Jun 2021 Soumya Ghosh, Q. Vera Liao, Karthikeyan Natesan Ramamurthy, Jiri Navratil, Prasanna Sattigeri, Kush R. Varshney, Yunfeng Zhang

In this paper, we describe an open source Python toolkit named Uncertainty Quantification 360 (UQ360) for the uncertainty quantification of AI models.

Fairness

Finding the Homology of Decision Boundaries with Active Learning

1 code implementation NeurIPS 2020 Weizhi Li, Gautam Dasarathy, Karthikeyan Natesan Ramamurthy, Visar Berisha

We theoretically analyze the proposed framework and show that the query complexity of our active learning algorithm depends naturally on the intrinsic complexity of the underlying manifold.

Active Learning Meta-Learning +2

Characterizing the Latent Space of Molecular Deep Generative Models with Persistent Homology Metrics

no code implementations NeurIPS Workshop TDA_and_Beyond 2020 Yair Schiff, Vijil Chenthamarakshan, Karthikeyan Natesan Ramamurthy, Payel Das

In this work, we propose a method for measuring how well the latent space of deep generative models is able to encode structural and chemical features of molecular datasets by correlating latent space metrics with metrics from the field of topological data analysis (TDA).

Topological Data Analysis

Bridging Mode Connectivity in Loss Landscapes and Adversarial Robustness

3 code implementations ICLR 2020 Pu Zhao, Pin-Yu Chen, Payel Das, Karthikeyan Natesan Ramamurthy, Xue Lin

In this work, we propose to employ mode connectivity in loss landscapes to study the adversarial robustness of deep neural networks, and provide novel methods for improving this robustness.

Adversarial Robustness

Model Agnostic Multilevel Explanations

no code implementations NeurIPS 2020 Karthikeyan Natesan Ramamurthy, Bhanukiran Vinzamuri, Yunfeng Zhang, Amit Dhurandhar

The method can also leverage side information, where users can specify points for which they may want the explanations to be similar.

Understanding racial bias in health using the Medical Expenditure Panel Survey data

no code implementations4 Nov 2019 Moninder Singh, Karthikeyan Natesan Ramamurthy

Over the years, several studies have demonstrated that there exist significant disparities in health indicators in the United States population across various groups.

PI-Net: A Deep Learning Approach to Extract Topological Persistence Images

1 code implementation5 Jun 2019 Anirudh Som, Hongjun Choi, Karthikeyan Natesan Ramamurthy, Matthew Buman, Pavan Turaga

To the best of our knowledge, we are the first to propose the use of deep learning for computing topological features directly from data.

Human Activity Recognition Image Classification +1

Teaching AI to Explain its Decisions Using Embeddings and Multi-Task Learning

no code implementations5 Jun 2019 Noel C. F. Codella, Michael Hind, Karthikeyan Natesan Ramamurthy, Murray Campbell, Amit Dhurandhar, Kush R. Varshney, Dennis Wei, Aleksandra Mojsilović

Using machine learning in high-stakes applications often requires predictions to be accompanied by explanations comprehensible to the domain user, who has ultimate responsibility for decisions and outcomes.

Multi-Task Learning

Optimized Score Transformation for Consistent Fair Classification

no code implementations31 May 2019 Dennis Wei, Karthikeyan Natesan Ramamurthy, Flavio du Pin Calmon

We derive a closed-form expression for the optimal transformed scores and a convex optimization problem for the transformation parameters.

Classification Fairness +1

Counting and Segmenting Sorghum Heads

no code implementations30 May 2019 Min-hwan Oh, Peder Olsen, Karthikeyan Natesan Ramamurthy

We also propose a novel instance segmentation algorithm using the estimated density map, to identify the individual panicles in the presence of occlusion.

Crowd Counting Instance Segmentation +1

Crowd Counting with Decomposed Uncertainty

no code implementations15 Mar 2019 Min-hwan Oh, Peder A. Olsen, Karthikeyan Natesan Ramamurthy

Uncertainty quantification accompanied by point estimation can lead to a more informed decision, and even improve the prediction quality.

Crowd Counting

Bias Mitigation Post-processing for Individual and Group Fairness

no code implementations14 Dec 2018 Pranay K. Lohia, Karthikeyan Natesan Ramamurthy, Manish Bhide, Diptikalyan Saha, Kush R. Varshney, Ruchir Puri

Whereas previous post-processing approaches for increasing the fairness of predictions of biased classifiers address only group fairness, we propose a method for increasing both individual and group fairness.

Fairness General Classification

TED: Teaching AI to Explain its Decisions

no code implementations12 Nov 2018 Michael Hind, Dennis Wei, Murray Campbell, Noel C. F. Codella, Amit Dhurandhar, Aleksandra Mojsilović, Karthikeyan Natesan Ramamurthy, Kush R. Varshney

Artificial intelligence systems are being increasingly deployed due to their potential to increase the efficiency, scale, consistency, fairness, and accuracy of decisions.

Fairness

Perturbation Robust Representations of Topological Persistence Diagrams

1 code implementation ECCV 2018 Anirudh Som, Kowshik Thopalli, Karthikeyan Natesan Ramamurthy, Vinay Venkataraman, Ankita Shukla, Pavan Turaga

However, persistence diagrams are multi-sets of points and hence it is not straightforward to fuse them with features used for contemporary machine learning tools like deep-nets.

Teaching Meaningful Explanations

no code implementations29 May 2018 Noel C. F. Codella, Michael Hind, Karthikeyan Natesan Ramamurthy, Murray Campbell, Amit Dhurandhar, Kush R. Varshney, Dennis Wei, Aleksandra Mojsilovic

The adoption of machine learning in high-stakes applications such as healthcare and law has lagged in part because predictions are not accompanied by explanations comprehensible to the domain user, who often holds the ultimate responsibility for decisions and outcomes.

Topological Data Analysis of Decision Boundaries with Application to Model Selection

1 code implementation25 May 2018 Karthikeyan Natesan Ramamurthy, Kush R. Varshney, Krishnan Mody

We propose the labeled \v{C}ech complex, the plain labeled Vietoris-Rips complex, and the locally scaled labeled Vietoris-Rips complex to perform persistent homology inference of decision boundaries in classification tasks.

General Classification Model Selection +1

Simultaneous Parameter Learning and Bi-Clustering for Multi-Response Models

no code implementations29 Apr 2018 Ming Yu, Karthikeyan Natesan Ramamurthy, Addie Thompson, Aurélie Lozano

We consider multi-response and multitask regression models, where the parameter matrix to be estimated is expected to have an unknown grouping structure.

Exploring High-Dimensional Structure via Axis-Aligned Decomposition of Linear Projections

no code implementations19 Dec 2017 Jayaraman J. Thiagarajan, Shusen Liu, Karthikeyan Natesan Ramamurthy, Peer-Timo Bremer

Furthermore, we introduce a new approach to discover a diverse set of high quality linear projections and show that in practice the information of $k$ linear projections is often jointly encoded in $\sim k$ axis aligned plots.

Distribution-Preserving k-Anonymity

no code implementations5 Nov 2017 Dennis Wei, Karthikeyan Natesan Ramamurthy, Kush R. Varshney

Preserving the privacy of individuals by protecting their sensitive attributes is an important consideration during microdata release.

Quantization Transfer Learning

Learning Robust Representations for Computer Vision

no code implementations31 Jul 2017 Peng Zheng, Aleksandr Y. Aravkin, Karthikeyan Natesan Ramamurthy, Jayaraman Jayaraman Thiagarajan

Unsupervised learning techniques in computer vision often require learning latent representations, such as low-dimensional linear and non-linear subspaces.

Representation Learning

Estimating Shape Parameters of Piecewise Linear-Quadratic Problems

no code implementations6 Jun 2017 Peng Zheng, Aleksandr Y. Aravkin, Karthikeyan Natesan Ramamurthy

The normalization constant inherent in this requirement helps to inform the optimization over shape parameters, giving a joint optimization problem over these as well as primary parameters of interest.

A Deep Learning Approach To Multiple Kernel Fusion

no code implementations28 Dec 2016 Huan Song, Jayaraman J. Thiagarajan, Prasanna Sattigeri, Karthikeyan Natesan Ramamurthy, Andreas Spanias

Kernel fusion is a popular and effective approach for combining multiple features that characterize different aspects of data.

Activity Recognition

Robust Local Scaling using Conditional Quantiles of Graph Similarities

no code implementations14 Dec 2016 Jayaraman J. Thiagarajan, Prasanna Sattigeri, Karthikeyan Natesan Ramamurthy, Bhavya Kailkhura

In this paper, we propose the use of quantile analysis to obtain local scale estimates for neighborhood graph construction.

graph construction

TreeView: Peeking into Deep Neural Networks Via Feature-Space Partitioning

no code implementations22 Nov 2016 Jayaraman J. Thiagarajan, Bhavya Kailkhura, Prasanna Sattigeri, Karthikeyan Natesan Ramamurthy

In this paper, we take a step in the direction of tackling the problem of interpretability without compromising the model accuracy.

A Riemannian Framework for Statistical Analysis of Topological Persistence Diagrams

1 code implementation28 May 2016 Rushil Anirudh, Vinay Venkataraman, Karthikeyan Natesan Ramamurthy, Pavan Turaga

This paper concerns itself with one popular topological feature, which is the number of $d-$dimensional holes in the dataset, also known as the Betti$-d$ number.

Topological Data Analysis

Persistent Homology of Attractors For Action Recognition

no code implementations16 Mar 2016 Vinay Venkataraman, Karthikeyan Natesan Ramamurthy, Pavan Turaga

In this paper, we propose a novel framework for dynamical analysis of human actions from 3D motion capture data using topological data analysis.

Action Recognition Time Series +1

Automatic Inference of the Quantile Parameter

no code implementations12 Nov 2015 Karthikeyan Natesan Ramamurthy, Aleksandr Y. Aravkin, Jayaraman J. Thiagarajan

However, loss functions such as quantile and quantile Huber generalize the symmetric $\ell_1$ and Huber losses to the asymmetric setting, for a fixed quantile parameter.

Adaptive As-Natural-As-Possible Image Stitching

no code implementations CVPR 2015 Chung-Ching Lin, Sharathchandra U. Pankanti, Karthikeyan Natesan Ramamurthy, Aleksandr Y. Aravkin

Computing the warp is fully automated and uses a combination of local homography and global similarity transformations, both of which are estimated with respect to the target.

Image Stitching

Beyond L2-Loss Functions for Learning Sparse Models

no code implementations26 Mar 2014 Karthikeyan Natesan Ramamurthy, Aleksandr Y. Aravkin, Jayaraman J. Thiagarajan

We propose an algorithm to learn dictionaries and obtain sparse codes when the data reconstruction fidelity is measured using any smooth PLQ cost function.

online learning Sparse Learning +1

Recovering Non-negative and Combined Sparse Representations

no code implementations12 Mar 2013 Karthikeyan Natesan Ramamurthy, Jayaraman J. Thiagarajan, Andreas Spanias

For case (c), we propose the combined orthogonal matching pursuit algorithm for coefficient recovery and derive the deterministic sparsity threshold under which recovery of the unique, sparsest coefficient vector is possible.

Learning Stable Multilevel Dictionaries for Sparse Representations

no code implementations3 Mar 2013 Jayaraman J. Thiagarajan, Karthikeyan Natesan Ramamurthy, Andreas Spanias

Algorithmic stability and generalization are desirable characteristics for dictionary learning algorithms that aim to build global dictionaries which can efficiently model any test data similar to the training samples.

Dictionary Learning

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