Acknowledgements

I am grateful to Edwige Cyffers and Yann Chevaleyre (coolest supervisors ever) for giving me the opportunity to pursue a PhD at Université Paris Dauphine–PSL, and for their guidance and support as my doctoral supervisors.

I am deeply grateful to my master’s thesis supervisor, Debabrota Basu (affectionately, Debu Da), for introducing me to the mathematical foundations of reinforcement learning, statistical learning theory, multi-armed bandits, algorithmic fairness, and performative learning.

I thank Sourav Chakraborty (best day or worst day: mood and homework decide) and Arijit Ghosh (very chill; rigorous maths, relentless roasts) at ACMU, ISI Kolkata, for giving me much-needed exposure to graph theory and theoretical computer science. Kuntal Ghosh (revisionist thinker) at MIU, ISI Kolkata, played a significant role in encouraging me to pursue research for the benefit of society.

My coursework at RKMVERI and ISI was enriched by many teachers:

I remain grateful for their teaching, guidance, and encouragement.