Filter Results:
(13)
Show Results For
- All HBS Web
(13)
- Research (11)
- Faculty Publications (8)
Show Results For
- All HBS Web
(13)
- Research (11)
- Faculty Publications (8)
Page 1 of 13
Results
- Article
Counterfactual Explanations Can Be Manipulated
By: Dylan Slack, Sophie Hilgard, Himabindu Lakkaraju and Sameer Singh
Counterfactual explanations are useful for both generating recourse and auditing fairness between groups. We seek to understand whether adversaries can manipulate counterfactual explanations in an algorithmic recourse setting: if counterfactual explanations indicate... View Details
Slack, Dylan, Sophie Hilgard, Himabindu Lakkaraju, and Sameer Singh. "Counterfactual Explanations Can Be Manipulated." Advances in Neural Information Processing Systems (NeurIPS) 34 (2021).
- 2022
- Article
Exploring Counterfactual Explanations Through the Lens of Adversarial Examples: A Theoretical and Empirical Analysis.
By: Martin Pawelczyk, Chirag Agarwal, Shalmali Joshi, Sohini Upadhyay and Himabindu Lakkaraju
As machine learning (ML) models become more widely deployed in high-stakes applications, counterfactual explanations have emerged as key tools for providing actionable model explanations in practice. Despite the growing popularity of counterfactual explanations, a... View Details
Keywords: Machine Learning Models; Counterfactual Explanations; Adversarial Examples; Mathematical Methods
Pawelczyk, Martin, Chirag Agarwal, Shalmali Joshi, Sohini Upadhyay, and Himabindu Lakkaraju. "Exploring Counterfactual Explanations Through the Lens of Adversarial Examples: A Theoretical and Empirical Analysis." Proceedings of the International Conference on Artificial Intelligence and Statistics (AISTATS) 25th (2022).
- Article
Learning Models for Actionable Recourse
By: Alexis Ross, Himabindu Lakkaraju and Osbert Bastani
As machine learning models are increasingly deployed in high-stakes domains such as legal and financial decision-making, there has been growing interest in post-hoc methods for generating counterfactual explanations. Such explanations provide individuals adversely... View Details
Ross, Alexis, Himabindu Lakkaraju, and Osbert Bastani. "Learning Models for Actionable Recourse." Advances in Neural Information Processing Systems (NeurIPS) 34 (2021).
- Research Summary
Selection, Reallocation, and Spillover: Identifying the Sources of Gains from Multinational Production (with Maggie Chen)
By: Laura Alfaro
Quantifying the gains from multinational production has been a vital topic of economic research. Positive productivity gains are often attributed to knowledge spillover from multinational to domestic firms. An alternative, less stressed explanation is firm selection... View Details
- 2023
- Working Paper
The Market for Sharing Interest Rate Risk: Quantities and Asset Prices
By: Umang Khetan, Jane Li, Ioana Neamtu and Ishita Sen
We study the extent of interest rate risk sharing across the financial system using granular positions and transactions data in interest rate swaps. We show that pension and insurance (PF&I) sector emerges as a natural counterparty to banks and corporations: overall,... View Details
Keywords: Interest Rates; Investment Funds; Banks and Banking; Insurance; Investment Banking; Risk and Uncertainty
Khetan, Umang, Jane Li, Ioana Neamtu, and Ishita Sen. "The Market for Sharing Interest Rate Risk: Quantities and Asset Prices." Harvard Business School Working Paper, No. 24-052, February 2024.
- Article
Beyond Individualized Recourse: Interpretable and Interactive Summaries of Actionable Recourses
By: Kaivalya Rawal and Himabindu Lakkaraju
As predictive models are increasingly being deployed in high-stakes decision-making, there has been a lot of interest in developing algorithms which can provide recourses to affected individuals. While developing such tools is important, it is even more critical to... View Details
Rawal, Kaivalya, and Himabindu Lakkaraju. "Beyond Individualized Recourse: Interpretable and Interactive Summaries of Actionable Recourses." Advances in Neural Information Processing Systems (NeurIPS) 33 (2020).
- TeachingInterests
Interpretability and Explainability in Machine Learning
As machine learning models are increasingly being employed to aid decision makers in high-stakes settings such as healthcare and criminal justice, it is important to ensure that the decision makers correctly understand and consequent trust the functionality of these... View Details
- Web
Research - Behavioral Finance & Financial Stability
fintech platform, the authors find that, compared to actual outcomes of the fintech platform’s model, the counterfactual outcomes based on a “traditional model” used for regulatory reporting purposes would result in a 60% higher... View Details
- 20 Jan 2010
- First Look
First Look: Jan. 20
globalization and liberalization have increased the competitiveness of product markets, one explanation for the trend towards decentralization could be increased competition. Of course there are a range of other factors that may also be... View Details
Keywords: Martha Lagace
- 27 Jun 2017
- First Look
First Look at New Research and Ideas, June 27
ambiguous effect on high-quality output. To evaluate this trade-off, I develop a procedure to estimate agents' effort costs and simulate counterfactuals under alternative feedback policies. The results suggest that feedback on net... View Details
Keywords: Sean Silverthorne
- 09 May 2017
- First Look
New Research and Ideas, May 9
the data support our model over alternative explanations such as recession-induced reduction in agency costs (due to managerial fears of bankruptcy) and changing coordination costs. Countries with more decentralized firms (like the U.S.)... View Details
Keywords: Sean Silverthorne
- 17 Jul 2012
- First Look
First Look: July 17
gains from multinational production has been a vital topic of economic research. Positive productivity gains are often attributed to knowledge spillover from multinational to domestic firms. An alternative, less stressed explanation is... View Details
Keywords: Sean Silverthorne