Bernard CVaR-UCBVI is minimax-optimal over the full return-law class in the leading-order regime; the lower-order terms retain their $\tau^{-1}$ dependence.
Abstract
For finite-horizon tabular CVaR reinforcement learning, prior work proves a $\widetilde{O}(\tau^{-1}\sqrt{SAK})$ leading regret bound for arbitrary normalized return laws and the sharper $\widetilde{O}(\sqrt{SAK/\tau})$ rate under a density lower bound. We show that the same Bernstein CVaR-UCBVI algorithm attains the sharper rate without continuity assumptions. The key is a selected-budget self-bound: the conditional variance of the episode shortfall is at most $\tau$ plus the value-estimation width. Substitution into the original Bernstein decomposition yields, with high probability, $\widetilde{O}(\sqrt{SAK/\tau}+(SAHK^{1/4}+S^2AH)/\tau)$ regret for arbitrary normalized return laws, including atomic, mixed, and continuous laws. The $\tau^{-1/2}$ leading term matches the expected-regret minimax lower bound up to logarithmic factors. Thus Bernstein CVaR-UCBVI is minimax-optimal over the full return-law class in the leading-order regime; the lower-order terms retain their $\tau^{-1}$ dependence.
We study the time-uniform convergence of the raw iterate of standard stochastic gradient descent (SGD) for unconstrained smooth convex objectives. We prove that, under standard noise assumptions, the time-uniform convergence rate gets arbitrarily close to $\sqrt{\log n / n}$ but never reaches it. More specifically, we...
Let $\pi(\mathrm{d} x)\propto e^{-U(x)}\, \mathrm{d} x$ on $\mathbb{R}^d$, where $U$ is continuously differentiable and $m$-strongly convex with a globally $L$-Lipschitz gradient, $0<m\leq L<\infty$, and $\kappa=L/m$. Fixed-step Metropolis-adjusted Langevin algorithm (MALA) has known warm-start mixing-time upper bounds...
Can the classical Heavy-Ball method, with arbitrary horizon-dependent parameters chosen in advance, achieve Nesterov's $O(T^{-2})$ last-iterate rate on every smooth convex objective? We provide a negative answer. For every horizon $T\ge2$ and every predetermined schedule with nonnegative step sizes and momenta in $[0,1...
We establish near-linear accuracy bounds for the classical Moreau--Yosida unadjusted Langevin algorithm (MYULA). The target is $\pi\propto e^{-f-g}$, where $f\in C^2(\mathbb{R}^d)$ is $m$-strongly convex with Lipschitz gradient and $g$ is convex and globally Lipschitz. Under an explicit parameter-dependent step-size co...
We consider the problem of noisy gradient-free minimization of the k-th order partial derivative of a $\beta$-H{\"o}lder function supported on a d-dimensional cube. We show that T ^{($\beta$+d+k)/(2$\beta$+d)} log(T )^{(\beta-k)/(2\beta+d)} is a non-asymptotic minimax rate of the T step cumulative regret for all $\beta...
Théo Paquier, A. Tsybakov, F. Portier et al.· 0 citations
Bakhtiari, Lattimore and Szepesv\'ari (COLT 2025) proved that Thompson sampling (TS) has Bayesian regret $\tilde O(d^{5/2}\sqrt n)$ for bandit convex optimisation with convex \emph{monotone} ridge losses $f(x)=\ell(\ip{x}{\theta})$, and asked whether monotonicity of the link is necessary. We give a qualitative negative...
Xuan Li· 0 citations
Related blog posts
MIT News · Artificial Intelligence· news.mit.eduOct 23, 2026
Assistant Professor Benjamin Lindquist studies how engineers deliberately injected randomness into early AI, giving rise to deep-learning models that sometimes hallucinate.
