Searcharxiv⌕ Search

arXiv subjects

Giovanni Montanari

Publications and source records attributed to Giovanni Montanari.

3 recordsLinked to original sources

A short proof of the 1-1/e performance of RANKING on the complete upper triangular instance

The randomized online bipartite matching algorithm RANKING of Karp, Vazirani and Vazirani (1990) matches, asymptotically, a fraction $1-1/e$ of the vertices of the complete upper triangular (CUT) instance. We give a short proof of this fact that looks at the final matching from the offline side. A single observation drives the argument: the first $k$ arrivals can distinguish the offline vertices $v_k,\ldots,v_n$ only through their ranks. As a consequence, whether $v_k$ is matched is decided by a uniform draw among the remaining ranks, and the number of matched vertices among $v_1,\ldots,v_k$ is a Markov chain with explicit transition probabilities. A linear differential equation, solved by hand, gives the limiting profile of this chain, and the constant $1/e$ appears where the matching probability reaches one.

cs.DS↗

Learning to Allocate Resources with Censored Feedback

We study an online resource allocation problem in which at each round, a budget $B$ must be allocated across $K$ arms under censored feedback. An arm yields a reward if and only if two conditions are satisfied: (i) the arm is activated according to an arm-specific Bernoulli random variable with unknown parameter, and (ii) the allocated budget exceeds a random threshold drawn from a parametric distribution with unknown parameter. Over $T$ rounds, the learner must jointly estimate the unknown parameters and allocate the budget so as to maximize cumulative reward facing the exploration--exploitation trade-off. We prove a worst-case regret lower bound $Ω(\sqrt{T})$ for the general model, even with known uniform threshold distributions. We then propose RA-UCB, an optimistic algorithm that leverages non-trivial parameter estimation and confidence bounds. When the budget $B$ is known at the beginning of each round, RA-UCB achieves a regret bound of order $\widetilde{\mathcal{O}}(\sqrt{T})$, and even $\mathcal{O}(\mathrm{poly}\text{-}\log T)$ under stronger assumptions. As for unknown, round dependent budget, we introduce \texttt{MG-UCB}, which allows within-round switching and infinitesimal allocations, and matches the regret guarantees of RA-UCB. We then validate our theoretical results through experiments on real-world datasets.

cs.LG↗

Learning When to Automate: Queue Control in Human-AI Service Systems

We study a human-AI service system in which tasks arrive sequentially and are processed through a two-stage architecture: an automated chatbot followed, when necessary, by a human agent. We consider $T$ sequentially arriving tasks, each belonging to one of $K$ heterogeneous types. For each task the decision maker chooses how many resources to allocate to the chatbot, whose type-dependent success probabilities are initially unknown. Tasks not resolved by the chatbot enter type-dependent human-service queues, where they are processed by a human agent with unknown service rates. This model captures a central tradeoff in hybrid service systems: relying more on automation reduces human congestion but increases chatbot costs, while insufficient automation may overload the human agent. We propose the UCB-DPP policy, which combines Upper Confidence Bounds with Drift-Plus-Penalty control to learn the unknown parameters of the system while making queue-aware decisions. We prove that UCB-DPP achieves regret $\widetilde{\mathcal{O}}(K\sqrt{T})$ and guarantees mean-rate stability of the human-service queues. Simulations on synthetic instances show that the proposed policy outperforms natural baselines.

cs.LG↗