User profiles for A. Dhurandhar
Amit DhurandharPrincipal Research Scientist, IBM Verified email at us.ibm.com Cited by 3382 |
Explanations based on the missing: Towards contrastive explanations with pertinent negatives
In this paper we propose a novel method that provides contrastive explanations justifying the
classification of an input by a black box classifier such as a deep neural network. Given an …
classification of an input by a black box classifier such as a deep neural network. Given an …
Predicting human olfactory perception from chemical features of odor molecules
A Keller, RC Gerkin, Y Guan, A Dhurandhar, G Turu… - Science, 2017 - science.org
It is still not possible to predict whether a given molecule will have a perceived odor or what
olfactory percept it will produce. We therefore organized the crowd-sourced DREAM …
olfactory percept it will produce. We therefore organized the crowd-sourced DREAM …
[PDF][PDF] Human-centered explainability for life sciences, healthcare, and medical informatics
Rapid advances in artificial intelligence (AI) and availability of biological, medical, and
healthcare data have enabled the development of a wide variety of models. Significant success …
healthcare data have enabled the development of a wide variety of models. Significant success …
Invariant risk minimization games
…, K Varshney, A Dhurandhar - International …, 2020 - proceedings.mlr.press
The standard risk minimization paradigm of machine learning is brittle when operating in
environments whose test distributions are different from the training distribution due to spurious …
environments whose test distributions are different from the training distribution due to spurious …
One explanation does not fit all: A toolkit and taxonomy of ai explainability techniques
As artificial intelligence and machine learning algorithms make further inroads into society,
calls are increasing from multiple stakeholders for these algorithms to explain their outputs. At …
calls are increasing from multiple stakeholders for these algorithms to explain their outputs. At …
TED: Teaching AI to explain its decisions
Artificial intelligence systems are being increasingly deployed due to their potential to increase
the efficiency, scale, consistency, fairness, and accuracy of decisions. However, as many …
the efficiency, scale, consistency, fairness, and accuracy of decisions. However, as many …
Ai explainability 360 toolkit
As machine learning algorithms make inroads into our lives and society, calls are increasing
from multiple stakeholders for these algorithms to explain their outputs. Moreover, these …
from multiple stakeholders for these algorithms to explain their outputs. Moreover, these …
Efficient data representation by selecting prototypes with importance weights
KS Gurumoorthy, A Dhurandhar… - … Conference on Data …, 2019 - ieeexplore.ieee.org
Prototypical examples that best summarize and compactly represent an underlying complex
data distribution, communicate meaningful insights to humans in domains where simple …
data distribution, communicate meaningful insights to humans in domains where simple …
Deciding fast and slow: The role of cognitive biases in AI-assisted decision-making
Several strands of research have aimed to bridge the gap between artificial intelligence (AI)
and human decision-makers in AI-assisted decision-making, where humans are the …
and human decision-makers in AI-assisted decision-making, where humans are the …
Model agnostic contrastive explanations for structured data
Recently, a method [7] was proposed to generate contrastive explanations for differentiable
models such as deep neural networks, where one has complete access to the model. In this …
models such as deep neural networks, where one has complete access to the model. In this …