arXiv subjects

Viswanath Nagarajan

Publications and source records attributed to Viswanath Nagarajan.

At least 19 recordsLinked to original sources

arXiv2026

Semi-Bandit Learning for Monotone Stochastic Optimization

Stochastic optimization is a widely used approach for optimization under uncertainty, where uncertain input parameters are modeled by random variables. Exact or approximation algorithms have been obtained for several fundamental problems in this area. However, a significant limitation of this approach is that it requires full knowledge of the underlying probability distributions. Can we still get good (approximation) algorithms if these distributions are unknown, and the algorithm needs to learn them through repeated interactions? In this paper, we resolve this question for a large class of ''monotone'' stochastic problems, by providing a generic online learning algorithm with $\sqrt{T\log(T)}$ regret relative to the best approximation algorithm (under known distributions). Importantly, our online algorithm works in a semi-bandit setting, where in each period, the algorithm only observes samples from the random variables that were actually probed. Moreover, our result extends to settings with censored and binary feedback, where the policy only observes truncated or thresholded versions of the probed variables. Our framework applies to several fundamental problems such as prophet inequality, Pandora's box, stochastic knapsack, single-resource revenue management and sequential posted pricing.

cs.LG↗

arXiv2026

Distributionally Robust $k$-of-$n$ Sequential Testing

The $k$-of-$n$ testing problem involves performing $n$ independent tests sequentially, in order to determine whether/not at least $k$ tests pass. The objective is to minimize the expected cost of testing. This is a fundamental and well-studied stochastic optimization problem. However, a key limitation of this model is that the success/failure probability of each test is assumed to be known precisely. In this paper, we relax this assumption and study a distributionally-robust model for $k$-of-$n$ testing. In our setting, each test is associated with an interval that contains its (unknown) failure probability. The goal is to find a solution that minimizes the worst-case expected cost, where each test's probability is chosen from its interval. We focus on non-adaptive solutions, that are specified by a fixed permutation of the tests. When all test costs are unit, we obtain a $2$-approximation algorithm for distributionally-robust $k$-of-$n$ testing. For general costs, we obtain an $O(\frac{1}{\sqrt ε})$-approximation algorithm on $ε$-bounded instances where each uncertainty interval is contained in $[ε, 1-ε]$. We also consider the inner maximization problem for distributionally-robust $k$-of-$n$: this involves finding the worst-case probabilities from the uncertainty intervals for a given solution. For this problem, in addition to the above approximation ratios, we obtain a quasi-polynomial time approximation scheme under the assumption that all costs are polynomially bounded.

cs.DS↗

arXiv2026

Learning Markov Decision Processes under Fully Bandit Feedback

A standard assumption in Reinforcement Learning is that the agent observes every visited state-action pair in the associated Markov Decision Process (MDP), along with the per-step rewards. Strong theoretical results are known in this setting, achieving nearly-tight $Θ(\sqrt{T})$-regret bounds. However, such detailed feedback can be unrealistic, and recent research has investigated more restricted settings such as trajectory feedback, where the agent observes all the visited state-action pairs, but only a single \emph{aggregate} reward. In this paper, we consider a far more restrictive ``fully bandit'' feedback model for episodic MDPs, where the agent does not even observe the visited state-action pairs -- it only learns the aggregate reward. We provide the first efficient bandit learning algorithm for episodic MDPs with $\widetilde{O}(\sqrt{T})$ regret. Our regret has an exponential dependence on the horizon length $\H$, which we show is necessary. We also obtain improved nearly-tight regret bounds for ``ordered'' MDPs; these can be used to model classical stochastic optimization problems such as $k$-item prophet inequality and sequential posted pricing. Finally, we evaluate the empirical performance of our algorithm for the setting of $k$-item prophet inequalities; despite the highly restricted feedback, our algorithm's performance is comparable to that of a state-of-art learning algorithm (UCB-VI) with detailed state-action feedback.

cs.LG↗

arXiv2025

A Simple Approximation Algorithm for Optimal Decision Tree

Optimal decision tree (\odt) is a fundamental problem arising in applications such as active learning, entity identification, and medical diagnosis. An instance of \odt is given by $m$ hypotheses, out of which an unknown ``true'' hypothesis is drawn according to some probability distribution. An algorithm needs to identify the true hypothesis by making queries: each query incurs a cost and has a known response for each hypothesis. The goal is to minimize the expected query cost to identify the true hypothesis. We consider the most general setting with arbitrary costs, probabilities and responses. \odt is NP-hard to approximate better than $\ln m$ and there are $O(\ln m)$ approximation algorithms known for it. However, these algorithms and/or their analyses are quite complex. Moreover, the leading constant factors are large. We provide a simple algorithm and analysis for \odt, proving an approximation ratio of $8 \ln m$.

cs.DS↗

arXiv2025

Identifying Approximate Minimizers under Stochastic Uncertainty

We study a fundamental stochastic selection problem involving $n$ independent random variables, each of which can be queried at some cost. Given a tolerance level $δ$, the goal is to find a value that is $δ$-approximately minimum (or maximum) over all the random variables, at minimum expected cost. A solution to this problem is an adaptive sequence of queries, where the choice of the next query may depend on previously-observed values. Two variants arise, depending on whether the goal is to find a $δ$-minimum value or a $δ$-minimizer. When all query costs are uniform, we provide a $4$-approximation algorithm for both variants. When query costs are non-uniform, we provide a $5.83$-approximation algorithm for the $δ$-minimum value and a $7.47$-approximation for the $δ$-minimizer. All our algorithms rely on non-adaptive policies (that perform a fixed sequence of queries), so we also upper bound the corresponding ''adaptivity'' gaps. Our analysis relates the stopping probabilities in the algorithm and optimal policies, where a key step is in proving and using certain stochastic dominance properties.

cs.DS↗

arXiv2025

Sequential Testing with Subadditive Costs

In the classic sequential testing problem, we are given a system with several components each of which fails with some independent probability. The goal is to identify whether or not some component has failed. When the test costs are additive, it is well known that a greedy algorithm finds an optimal solution. We consider a much more general setting with subadditive cost functions and provide a $(4ρ+γ)$-approximation algorithm, assuming a $γ$-approximate value oracle (that computes the cost of any subset) and a $ρ$-approximate ratio oracle (that finds a subset with minimum ratio of cost to failure probability). While the natural greedy algorithm has a poor approximation ratio in the subadditive case, we show that a suitable truncation achieves the above guarantee. Our analysis is based on a connection to the minimum sum set cover problem. As applications, we obtain the first approximation algorithms for sequential testing under various cost-structures: $(5+ε)$-approximation for tree-based costs, $9.5$-approximation for routing costs and $(4+\ln n)$ for machine activation costs. We also show that sequential testing under submodular costs does not admit any poly-logarithmic approximation (assuming the exponential time hypothesis).

cs.DS↗

arXiv2024

Optimal Decision Tree and Adaptive Submodular Ranking with Noisy Outcomes

In pool-based active learning, the learner is given an unlabeled data set and aims to efficiently learn the unknown hypothesis by querying the labels of the data points. This can be formulated as the classical Optimal Decision Tree (ODT) problem: Given a set of tests, a set of hypotheses, and an outcome for each pair of test and hypothesis, our objective is to find a low-cost testing procedure (i.e., decision tree) that identifies the true hypothesis. This optimization problem has been extensively studied under the assumption that each test generates a deterministic outcome. However, in numerous applications, for example, clinical trials, the outcomes may be uncertain, which renders the ideas from the deterministic setting invalid. In this work, we study a fundamental variant of the ODT problem in which some test outcomes are noisy, even in the more general case where the noise is persistent, i.e., repeating a test gives the same noisy output. Our approximation algorithms provide guarantees that are nearly best possible and hold for the general case of a large number of noisy outcomes per test or per hypothesis where the performance degrades continuously with this number. We numerically evaluated our algorithms for identifying toxic chemicals and learning linear classifiers, and observed that our algorithms have costs very close to the information-theoretic minimum.

cs.LG↗

arXiv2024

Lower Bound on the Greedy Approximation Ratio for Adaptive Submodular Cover

We show that the greedy algorithm for adaptive-submodular cover has approximation ratio at least 1.3*(1+ln Q). Moreover, the instance demonstrating this gap has Q=1. So, it invalidates a prior result in the paper ``Adaptive Submodularity: A New Approach to Active Learning and Stochastic Optimization'' by Golovin-Krause, that claimed a (1+ln Q)^2 approximation ratio for the same algorithm.

cs.DS↗

arXiv2024

Minimum Cost Adaptive Submodular Cover

Adaptive submodularity is a fundamental concept in stochastic optimization, with numerous applications such as sensor placement, hypothesis identification and viral marketing. We consider the problem of minimum cost cover of adaptive-submodular functions, and provide a $4(1+\ln Q)$-approximation algorithm, where $Q$ is the goal value. In fact, we consider a significantly more general objective of minimizing the $p^{th}$ moment of the coverage cost, and show that our algorithm simultaneously achieves a $(p+1)^{p+1}\cdot (\ln Q+1)^p$ approximation guarantee for all $p\ge 1$. All our approximation ratios are best possible up to constant factors (assuming $P\ne NP$). Moreover, our results also extend to the setting where one wants to cover {\em multiple} adaptive-submodular functions. Finally, we evaluate the empirical performance of our algorithm on instances of hypothesis identification.

cs.DS↗

arXiv2024

Cluster Before You Hallucinate: Approximating Node-Capacitated Network Design and Energy Efficient Routing

We consider the following node-capacitated network design problem. The input is an undirected graph, set of demands, uniform node capacity and arbitrary node costs. The goal is to find a minimum node-cost subgraph that supports all demands concurrently subject to the node capacities. We consider both single and multi-commodity demands, and provide the first poly-logarithmic approximation guarantees. For single-commodity demands (i.e., all request pairs have the same sink node), we obtain an $O(\log^2 n)$ approximation to the cost with an $O(\log^3 n)$ factor violation in node capacities. For multi-commodity demands, we obtain an $O(\log^4 n)$ approximation to the cost with an $O(\log^{10} n)$ factor violation in node capacities. We use a variety of techniques, including single-sink confluent flows, low-load set cover, random sampling and cut-sparsification. We also develop new techniques for clustering multicommodity demands into (nearly) node-disjoint clusters, which may be of independent interest. Moreover, this network design problem has applications to energy-efficient virtual circuit routing. In this setting, there is a network of routers that are speed scalable, and that may be shutdown when idle. We assume the standard model for power: the power consumed by a router with load (speed) $s$ is $σ+ s^α$ where $σ$ is the static power and the exponent $α> 1$. We obtain the first poly-logarithmic approximation algorithms for this problem when speed-scaling occurs on nodes of a network.

cs.DS↗

arXiv2023

Informative Path Planning with Limited Adaptivity

We consider the informative path planning ($\mathtt{IPP}$) problem in which a robot interacts with an uncertain environment and gathers information by visiting locations. The goal is to minimize its expected travel cost to cover a given submodular function. Adaptive solutions, where the robot incorporates all available information to select the next location to visit, achieve the best objective. However, such a solution is resource-intensive as it entails recomputing after every visited location. A more practical approach is to design solutions with a small number of adaptive "rounds", where the robot recomputes only once at the start of each round. In this paper, we design an algorithm for $\mathtt{IPP}$ parameterized by the number $k$ of adaptive rounds, and prove a smooth trade-off between $k$ and the solution quality (relative to fully adaptive solutions). We validate our theoretical results by experiments on a real road network, where we observe that a few rounds of adaptivity suffice to obtain solutions of cost almost as good as fully-adaptive ones.

cs.DS↗

arXiv2023

Non-Adaptive Stochastic Score Classification and Explainable Halfspace Evaluation

Sequential testing problems involve a complex system with several components, each of which is "working" with some independent probability. The outcome of each component can be determined by performing a test, which incurs some cost. The overall system status is given by a function $f$ of the outcomes of its components. The goal is to evaluate this function $f$ by performing tests at the minimum expected cost. While there has been extensive prior work on this topic, provable approximation bounds are mainly limited to simple functions like ``k-out-of-n'' and halfspaces. We consider significantly more general "score classification" functions, and provide the first constant factor approximation algorithm (improving over a previous logarithmic approximation ratio). Moreover, our policy is non adaptive: it just involves performing tests in an a priori fixed order. We also consider the related halfspace evaluation problem, where we want to evaluate some function on $d$ halfspaces (e.g., intersection of halfspaces). We show that our approach provides an $O(d^2\log d)$-approximation algorithm for this problem. Our algorithms also extend to the setting of "batched'' tests, where multiple tests can be performed simultaneously while incurring an extra setup cost. Finally, we perform computational experiments that demonstrate the practical performance of our algorithm for score classification. We observe that, for most instances, the cost of our algorithm is within $50\%$ of an information-theoretic lower bound on the optimal value.

cs.DS↗

arXiv2022

An Asymptotically Optimal Batched Algorithm for the Dueling Bandit Problem

We study the $K$-armed dueling bandit problem, a variation of the traditional multi-armed bandit problem in which feedback is obtained in the form of pairwise comparisons. Previous learning algorithms have focused on the $\textit{fully adaptive}$ setting, where the algorithm can make updates after every comparison. The "batched" dueling bandit problem is motivated by large-scale applications like web search ranking and recommendation systems, where performing sequential updates may be infeasible. In this work, we ask: $\textit{is there a solution using only a few adaptive rounds that matches the asymptotic regret bounds of the best sequential algorithms for $K$-armed dueling bandits?}$ We answer this in the affirmative $\textit{under the Condorcet condition}$, a standard setting of the $K$-armed dueling bandit problem. We obtain asymptotic regret of $O(K^2\log^2(K)) + O(K\log(T))$ in $O(\log(T))$ rounds, where $T$ is the time horizon. Our regret bounds nearly match the best regret bounds known in the fully sequential setting under the Condorcet condition. Finally, in computational experiments over a variety of real-world datasets, we observe that our algorithm using $O(\log(T))$ rounds achieves almost the same performance as fully sequential algorithms (that use $T$ rounds).

cs.LG↗

arXiv2022

Efficient Algorithms for Stochastic Ridepooling Assignment with Mixed Fleets

Ride-pooling, which accommodates multiple passenger requests in a single trip, has the potential to significantly increase fleet utilization in shared mobility platforms. The ride-pooling assignment problem finds optimal co-riders to maximize the total utility or profit on a shareability graph, a hypergraph representing the matching compatibility between available vehicles and pending requests. With mixed fleets due to the introduction of automated or premium vehicles, fleet sizing and relocation decisions should be made before the requests are revealed. Due to the immense size of the underlying shareability graph and demand uncertainty, it is impractical to use exact methods to calculate the optimal trip assignments. Two approximation algorithms for mid-capacity and high-capacity vehicles are proposed in this paper; The respective approximation ratios are $\frac1{p^2}$ and $\frac{e-1}{(2e+o(1)) p \ln p}$, where $p$ is the maximum vehicle capacity plus one. The performance of these algorithms is validated using a mixed autonomy on-demand mobility simulator. These efficient algorithms serve as a stepping stone for a variety of multimodal and multiclass on-demand mobility applications.

math.OC↗

arXiv2022

Batched Dueling Bandits

The $K$-armed dueling bandit problem, where the feedback is in the form of noisy pairwise comparisons, has been widely studied. Previous works have only focused on the sequential setting where the policy adapts after every comparison. However, in many applications such as search ranking and recommendation systems, it is preferable to perform comparisons in a limited number of parallel batches. We study the batched $K$-armed dueling bandit problem under two standard settings: (i) existence of a Condorcet winner, and (ii) strong stochastic transitivity and stochastic triangle inequality. For both settings, we obtain algorithms with a smooth trade-off between the number of batches and regret. Our regret bounds match the best known sequential regret bounds (up to poly-logarithmic factors), using only a logarithmic number of batches. We complement our regret analysis with a nearly-matching lower bound. Finally, we also validate our theoretical results via experiments on synthetic and real data.

cs.LG↗

arXiv2021

On Some Variants of Euclidean K-Supplier

The $k$-Supplier problem is an important location problem that has been actively studied in both general and Euclidean metrics. Many of its variants have also been studied, primarily on general metrics. We study two variants of $k$-Supplier, namely Priority $k$-Supplier and $k$-Supplier with Outliers, in Euclidean metrics. We obtain $(1+\sqrt{3})$-approximation algorithms for both variants, which are the first improvements over the previously-known factor-$3$ approximation (that is known to be best-possible for general metrics). We also study the Matroid Supplier problem on Euclidean metrics, and show that it cannot be approximated to a factor better than $3$ (assuming $P\ne NP$); so the Euclidean metric offers no improvement in this case.

cs.DS↗

arXiv2021

The Power of Adaptivity for Stochastic Submodular Cover

In the stochastic submodular cover problem, the goal is to select a subset of stochastic items of minimum expected cost to cover a submodular function. Solutions in this setting correspond to sequential decision processes that select items one by one "adaptively" (depending on prior observations). While such adaptive solutions achieve the best objective, the inherently sequential nature makes them undesirable in many applications. We ask: how well can solutions with only a few adaptive rounds approximate fully-adaptive solutions? We give nearly tight answers for both independent and correlated settings, proving smooth tradeoffs between the number of adaptive rounds and the solution quality, relative to fully adaptive solutions. Experiments on synthetic and real datasets show qualitative improvements in the solutions as we allow more rounds of adaptivity; in practice, solutions with a few rounds of adaptivity are nearly as good as fully adaptive solutions.

cs.DS↗

arXiv2021

Stochastic Makespan Minimization in Structured Set Systems

We study stochastic combinatorial optimization problems where the objective is to minimize the expected maximum load (a.k.a.\ the makespan). In this framework, we have a set of $n$ tasks and $m$ resources, where each task $j$ uses some subset of the resources. Tasks have random sizes $X_j$, and our goal is to non-adaptively select $t$ tasks to minimize the expected maximum load over all resources, where the load on any resource $i$ is the total size of all selected tasks that use $i$. For example, when resources are points and tasks are intervals in a line, we obtain an $O(\log\log m)$-approximation algorithm. Our technique is also applicable to other problems with some geometric structure in the relation between tasks and resources; e.g., packing paths, rectangles, and "fat" objects. Our approach uses a strong LP relaxation using the cumulant generating functions of the random variables. We also show that this LP has an $Ω(\log^* m)$ integrality gap, even for the problem of selecting intervals on a line; here $\log^* m$ is the iterated logarithm function.

cs.DS↗