Projects

Empirical Analysis of Computational Methods (EACM)

Link Prediction—Non-Uniform Missing Edges

Link Prediction Accuracy on Real-World Networks Under Non-Uniform Missing-Edge Patterns

Link Prediction—Sequential Stacking

Sequential Stacking Link Prediction

Link Prediction—Stacking Models

Stacking Models for Nearly Optimal Link Prediction in Complex Networks

Community Detection—Network Community Structure

Evaluating Overfit and Underfit in Models of Network Community Structure

Theoretical Analysis of Inference Limits (TAIL)

DOA Estimation in DS-CDMA

Subspace Based DOA Estimation of DS-CDMA Signals

Community Detection—Community Detectability

Detectability Thresholds and Optimal Algorithms for Community Structure in Dynamic Networks

Real-world Applications of Inference and Learning (RAIL)

AI Auditing—LLM-Assisted Reranking

LLM-Assisted Reranking to Operationalize Nuanced Objectives in Recommender Systems

AI Auditing—Seeking Help, Facing Harm

Auditing TikTok's Mental Health Recommendations

Antagonistic Ties—Structure and Function

The Structure and Function of Antagonistic Ties in Village Social Networks

AI Auditing—Counterfactual Bots

Causally Estimating the Effect of YouTube's Recommender System Using Counterfactual Bots

Antagonistic Ties—The Enmity Paradox

The enmity paradox: A person’s enemies have more enemies, on average, than a person does.

AI Auditing—Radical Content on YouTube

Examining the Consumption of Radical Content on YouTube