NISHEETH K. VISHNOI

RESEARCH

My work spans various areas of Mathematics, Theoretical Computer Science, Optimization, and Artificial Intelligence. I aim to tackle some of the most pressing and complex problems at the intersection of computation and society.

I study foundational questions about algorithmic fairness, privacy, and decision-making, especially in settings where algorithms interact with human judgment, institutional processes, and social norms. My work includes models of bias and strategic behavior in selection systems, as well as the design of equitable and private mechanisms. I also develop mathematical tools for efficient learning in diffusion models, particularly in geometrically structured spaces. More recently, I have been building theoretical frameworks to understand the impact of AI - such as large language models - on work, science, knowledge, and societal systems. This includes examining how AI alters skill formation, decision structures, and human-AI collaboration, and how we might build more accountable, interpretable, and humane computational systems in response.

At Yale, I co-founded the Computation and Society Initiative.

I am co-PI of an NSF funded AI Institute: The Institute for Learning-enabled Optimization at Scale

At Yale, I am affiliated to the Cowles Foundation for Research in Economics, the Institution for Social and Policy Studies, and the Thurman Arnold Project at the Yale School of Management.

I served on the Yale AI Task Force.

My Curriculum Vitae

Explore my essays centered around the question:
What does it mean to be intelligent in the age of AI?

When Agency Slips Away: The erosion of agency when fluency replaces reflection.
The Myth of Superintelligence: Why metrics and hype mislead our view of intelligence.
What Counts as Discovery?: On science, reframing, and the limits of AI prediction.
The Anatomy of Work in the Age of AI: How action and judgment diverge in tasks.
AI and the Erosion of Knowing: How answer machines hollow out curiosity.
What is Intelligence? Layers of Emergence: Intelligence as an emergent capacity.
What is Intelligence? Architecture, Divergence, and Fictions: The faculties of intelligence and the selves we build.

Subscribe to The Intelligence Loop to follow the full series.

AFFILIATIONS

CURRENT A. Bartlett Giamatti Professor of Computer Science, Yale

ADJUNCT IIT Kanpur

    PAST         IIT Bombay    Georgia Tech    IBM Research    UC Berkeley
                     CNRS    Microsoft Research    Simons Inst.   EPFL    IIT Delhi
                     ICTS Bangalore    IIT Goa    Google Research

OPEN POSITIONS

Several open positions Ph.D./Postdoc positions in theoretical computer science, machine learning, optimization, and algorithmic fairness are available. Please get in touch directly if you would be interested.

PUBLICATIONS

ALL PAPERS

RESEARCH ON ALGORITHMIC BIAS

Controlling Bias in AI

BOOKS and SURVEYS

    Algorithms for Convex Optimization
    Lx=b
    Optimization, Sampling, Lie Theory
    Hamiltonian Monte Carlo
    Geodesic Convex Optimization
    Faster Algorithms via Approximation Theory
    Real Stable Polynomials and their Applications to TCS
    Evolution