Multi-armed bandit algorithms are fundamental tools for sequential decision-making under uncertainty, with widespread applications across domains such as clinical trials and personalized decision-making. As bandit algorithms are increasingly deployed in these socially sensitive settings, it becomes critical to protect user data privacy and ensure fair treatment across decision rounds. While prior work has independently addressed privacy and fairness in bandit settings, the question of whether both objectives can be achieved simultaneously has remained largely open. Existing privacy-preserving bandit algorithms typically optimize average regret, a utilitarian measure, whereas fairness-aware approaches focus on minimizing Nash regret, which penalizes inequitable reward distributions, but often disregard privacy concerns. To bridge this gap, we introduce Differentially Private Nash Confidence Bound (DP-NCB)—a novel and unified algorithmic framework that simultaneously ensures ϵ-differential privacy and achieves order-optimal Nash regret, matching known lower bounds up to logarithmic factors. The framework is sufficiently general to operate under both global and local differential privacy models, and is anytime, requiring no prior knowledge of the time horizon. We support our theoretical guarantees with simulations on synthetic bandit instances, showing that DP-NCB incurs substantially lower Nash regret than state-of-the-art baselines. Our results offer a principled foundation for designing bandit algorithms that are both privacy-preserving and fair, making them suitable for high-stakes, socially impactful applications.

Vinay Shashidhar, Nishant Pandey, Varun Aggarwal. Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2015.

In this paper, we address the problem of grading spontaneous speech using a combination of machine learning and crowdsourcing. Traditional machine learning techniques solve the stated problem inadequately as automatic speaker-independent speech transcription is inaccurate. The features derived from it are also inaccurate and so is the machine learning model developed for speech evaluation. We propose a framework that combines machine learning with crowdsourcing. This entails identifying human intelligence tasks in the feature derivation step and using crowdsourcing to get them completed. We post the task of speech transcription to a large community of online workers (crowd). We also get spoken English grades from the crowd. We achieve 95% transcription accuracy by combining transcriptions from multiple crowd workers. Speech and prosody features are derived by force aligning the speech samples on these highly accurate transcriptions. Additionally, we derive surface and semantic level features directly from the transcription. We demonstrate the efficacy of our approach by predicting expert graded speech sample of 566 adult non-native speakers across two different countries - India and Philippines. Using the regression modeling technique, we are able achieve a Pearson correlation of 0.79 on the Philippines set and 0.74 on the Indian set with expert grades, an accuracy much higher than any previously reported machine learning approach. Our approach has an accuracy that rivals that of expert agreement. We show the value of the system through a case study in a real-world industrial deployment. This work is timely given the huge requirement of spoken English training and assessment.