Events

In the theory part of the talk, we study the statistical performance of maximum likelihood estimation (MLE) and, more generally, empirical risk minimization (ERM). While MLE is known to be minimax-optimal for low-complexity models, classical work showed that it can be suboptimal over “large” function classes, though the canonical examples are somewhat pathological. First, we develop a technique for detecting and quantifying the suboptimality of ERM in regression over high-dimensional nonparametric classes. Second, we show that the variance term of ERM procedures is always upper-bounded by the minimax rate, implying that any minimax suboptimality must arise from bias. Third, we present the first minimax-optimal estimator for convex regression in all dimensions with a polynomial runtime in the sample size. If time permits, we also discuss connections between the local theory of Banach spaces and minimum-norm interpolators, building on an approach initiated by Maurey and Pisier. In the applied part of the talk, we propose an explanation for the empirical success of test-time training (TTT) in foundation models, which we primarily validate through experiments with sparse autoencoders (SAEs). TTT identifies the training points most similar to a given evaluation point and improves predictions by locally adapting the model to this selected neighborhood. Although TTT has been studied previously, only recently has it been shown to deliver substantial gains in foundation models across domains such as image generation, control, and language modeling.

Financial systems are becoming increasingly digital and decentralized, demanding a practical fusion of distributed systems security and economic theory. A key enabler of this change, blockchain technology, promises more private and egalitarian economic mechanisms, built by facilitating consensus between pseudonymous actors. However, the theoretical security of these systems may mask significant real-world risks. In this talk, I will present recent advances in bridging this gap between theory and practice. First, I will discuss the resolution of a decade-old puzzle: the lack of observed attacks on major consensus mechanisms. I will then distill the lessons learnt into a holistic approach to designing robust mechanisms for distributed pseudonymous systems and demonstrate its adoption in practice using several lines of work.