About

Robert D. Kleinberg is a Professor of Computer Science at Cornell University, specializing in algorithms and theoretical computer science. His research focuses particularly on the economic aspects of algorithms, online learning and its applications, and random processes in networks. Throughout his career, he has contributed to advancing the understanding of algorithmic theory in these areas, exploring topics such as online convex optimization, learning in budgeted auctions, and the design of oblivious reconfigurable networks. Kleinberg has also been involved in teaching a variety of computer science courses at Cornell, including Introduction to Analysis of Algorithms, Mathematical Foundations of the Information Age, and Analysis of Algorithms, reflecting his commitment to education in theoretical computer science. His work is recognized for its depth and impact in both theoretical foundations and practical applications within computer science.

Research topics

• Computer Science
• Mathematics
• Combinatorics
• Sociology
• Optics
• Distributed computing
• Data science
• Statistics
• Chromatography
• Physics
• World Wide Web
• Econometrics
• Economics
• Mathematical analysis
• Computer network
• Applied mathematics
• Algorithm
• Discrete mathematics

Selected publications

• The \(k\)-Fold Matroid Secretary Problem

  Society for Industrial and Applied Mathematics eBooks · 2026-01-01

  book-chapter

  In the matroid secretary problem, elements \(N := [n]\) of a matroid \(\mathcal{M} \subseteq 2^{N}\) arrive in random order. When an element arrives, its weight is revealed and a choice must be made to accept or reject the element, subject to the constraint that the accepted set \(S \in \mathcal{M}\). [16] gives a \((1 - O(1/\sqrt k))\)-competitive algorithm when \(\mathcal{M}\) is \(k\)-uniform matroid. We generalize their result, giving a \((1 - O(\sqrt {\log(n)/k}))\)-competitive algorithm when \(\mathcal{M}\) is a \(k\)-fold matroid union.

  PublisherDOI

• Universal Connection Schedules for Reconfigurable Networking

  Society for Industrial and Applied Mathematics eBooks · 2026-01-01

  book-chapter

  Reconfigurable networks are a novel communication paradigm in which the pattern of connectivity between hosts varies rapidly over time. Prior theoretical work explored the inherent tradeoffs between throughput (or, hop-count) and latency, and showed the existence of infinitely many Pareto-optimal designs as the network size tends to infinity. Existing Pareto-optimal designs use a connection schedule which is fine-tuned to the desired hop-count \(h\), permitting lower latency as \(h\) increases. However, in reality datacenter workloads contain a mix of low-latency and high-latency requests. Using a connection schedule fine-tuned for one request type leads to inefficiencies when serving other types.

  PublisherDOI

• Full Swap Regret and Discretized Calibration

  ArXiv.org · 2025-02-13

  preprintOpen access

  We study the problem of minimizing swap regret in structured normal-form games. Players have a very large (potentially infinite) number of pure actions, but each action has an embedding into $d$-dimensional space and payoffs are given by bilinear functions of these embeddings. We provide an efficient learning algorithm for this setting that incurs at most $\tilde{O}(T^{(d+1)/(d+3)})$ swap regret after $T$ rounds. To achieve this, we introduce a new online learning problem we call \emph{full swap regret minimization}. In this problem, a learner repeatedly takes a (randomized) action in a bounded convex $d$-dimensional action set $\mathcal{K}$ and then receives a loss from the adversary, with the goal of minimizing their regret with respect to the \emph{worst-case} swap function mapping $\mathcal{K}$ to $\mathcal{K}$. For varied assumptions about the convexity and smoothness of the loss functions, we design algorithms with full swap regret bounds ranging from $O(T^{d/(d+2)})$ to $O(T^{(d+1)/(d+2)})$. Finally, we apply these tools to the problem of online forecasting to minimize calibration error, showing that several notions of calibration can be viewed as specific instances of full swap regret. In particular, we design efficient algorithms for online forecasting that guarantee at most $O(T^{1/3})$ $\ell_2$-calibration error and $O(\max(\sqrt{εT}, T^{1/3}))$ \emph{discretized-calibration} error (when the forecaster is restricted to predicting multiples of $ε$).

  PublisherOA PDFDOI

• Breaking the T^(2/3) Barrier for Sequential Calibration

  2025-06-15

  articleOpen access

  STOC ’25, Prague, Czechia

  PublisherOA PDFDOI

• Distributed Load Balancing with Workload-Dependent Service Rates

  2025-07-02

  articleOpen access

  In many real-world applications such as data centers and cloud computing, the systems often consist of multiple frontends (routers) that receive job requests and backends (servers) that process these jobs. Efficient resource management is becoming increasingly important given the growing demand for serving machine learning inference queries, which incur high latencies and require expensive computational resources.

  PublisherOA PDFDOI

• Near-Optimal Algorithms for Omniprediction

  ArXiv.org · 2025-01-28

  preprintOpen access

  Omnipredictors are simple prediction functions that encode loss-minimizing predictions with respect to a hypothesis class $H$, simultaneously for every loss function within a class of losses $L$. In this work, we give near-optimal learning algorithms for omniprediction, in both the online and offline settings. To begin, we give an oracle-efficient online learning algorithm that acheives $(L,H)$-omniprediction with $\tilde O (\sqrt{T \log |H|})$ regret for any class of Lipschitz loss functions $L \subseteq L_\mathrm{Lip}$. Quite surprisingly, this regret bound matches the optimal regret for \emph{minimization of a single loss function} (up to a $\sqrt{\log(T)}$ factor). Given this online algorithm, we develop an online-to-offline conversion that achieves near-optimal complexity across a number of measures. In particular, for all bounded loss functions within the class of Bounded Variation losses $L_\mathrm{BV}$ (which include all convex, all Lipschitz, and all proper losses) and any (possibly-infinite) $H$, we obtain an offline learning algorithm that, leveraging an (offline) ERM oracle and $m$ samples from $D$, returns an efficient $(L_{\mathrm{BV}},H,ε(m))$-omnipredictor for $\varepsilon(m)$ scaling near-linearly in the Rademacher complexity of a class derived from $H$ by taking convex combinations of a fixed number of elements of $\mathrm{Th} \circ H$.

  PublisherOA PDFDOI

• The Keychain Problem: On Minimizing the Opportunity Cost of Uncertainty

  arXiv (Cornell University) · 2025-09-07

  preprintOpen access

  In this paper, we introduce a family of sequential decision-making problems, collectively termed the Keychain Problem, that involve exploring a set of actions to maximize expected payoff when only a subset of actions are available in each stage. In an instance of the Keychain Problem, a locksmith faces a sequence of decisions, each of which involves selecting one key from a keychain (a subset of keys) to attempt to open a lock. Given a Bayesian prior on the effectiveness of keys, the locksmith's goal is to minimize the opportunity cost, which is the expected number of rounds in which the chain has a correct key but our selected key is incorrect. We study the computation of the Bayes optimal solution for Keychain Problems. Employing polynomial-time reductions, we establish formal connections between natural variants of the Keychain Problem and well-studied algorithmic economics problems on bipartite graphs. When the keychain order is known to the locksmith, we show that it reduces to Maximum Weight Bipartite Matching (MWBM). More general is the situation when the keychain order is sampled from a prior distribution (possibly correlated with the correct key). Here the Keychain Problem reduces to a novel generalization of MWBM which we coin the Maximum Weight Laminar Matching, which then further reduces to combinatorial auctions under XOS valuation functions. Finally, we show that when the locksmith can choose the keychain order, the Keychain problem reduces from a classic NP-hard combinatorial problem, again, on bipartite graphs. Besides implying algorithmic results and deepening our structural understanding about the Keychain Problem, our established reductions also find applications beyond -- for example, to the Philosopher Inequality for online bipartite matching.

  PublisherOA PDFDOI

• Universal Connection Schedules for Reconfigurable Networking

  ArXiv.org · 2025-11-11

  preprintOpen access

  Reconfigurable networks are a novel communication paradigm in which the pattern of connectivity between hosts varies rapidly over time. Prior theoretical work explored the inherent tradeoffs between throughput (or, hop-count) and latency, and showed the existence of infinitely many Pareto-optimal designs as the network size tends to infinity. Existing Pareto-optimal designs use a connection schedule which is fine-tuned to the desired hop-count $h$, permitting lower latency as $h$ increases. However, in reality datacenter workloads contain a mix of low-latency and high-latency requests. Using a connection schedule fine-tuned for one request type leads to inefficiencies when serving other types. A more flexible and efficient alternative is a {\em universal schedule}, a single connection schedule capable of attaining many Pareto-optimal tradeoff points simultaneously, merely by varying the choice of routing paths. In this work we present the first universal schedules for oblivious routing. Our constructions yield universal schedules which are near-optimal for all possible hop-counts $h$. The key technical idea is to specialize to a type of connection schedule based on cyclic permutations and to develop a novel Fourier-analytic method for analyzing randomized routing on these connection schedules. We first show that a uniformly random connection schedule suffices with multiplicative error in throughput, and latency optimal up to a $\log N$ factor. We then show that a more carefully designed random connection schedule suffices with additive error in throughput, but improved latency optimal up to only constant factors. Finally, we show that our first randomized construction can be made deterministic using a derandomized version of the Lovett-Meka discrepancy minimization algorithm to obtain the same result.

  PublisherOA PDFDOI

• SUSTAINABILITY IN ANAESTHESIA: PRESERVING THE PRIMACY OF PATIENT WELFARE WHILE REDUCING GREENHOUSE GAS EMISSIONS (revised)

  Journal of Cardiothoracic and Vascular Anesthesia · 2025-11-10

  article1st authorCorresponding
  PublisherDOI

• Learning in Budgeted Auctions with Spacing Objectives

  2025-07-02

  articleOpen access

  In this paper, we introduce a novel approach to repeated auctions that accounts for bidders' temporal preferences, important in applications such as advertising. In our model, when a player wins an auction after not winning for ℓ rounds, she is awarded r(ℓ) utility and her goal is to maximize her total utility. r : ℕ → ℝ ≥0 satisfies the following properties. (i) The more rounds without a win, the higher the reward, i.e., r is weakly increasing. (ii) As more rounds pass without winning, the increase in reward becomes smaller, i.e., r is concave. The motivation behind these properties comes from the advertising literature, which states that an increased frequency of winning builds advertising effectiveness at a decreasing (but not declining) rate. The above properties guarantee that adding more wins to any sequence of winning intervals increases the total reward.

  PublisherOA PDFDOI

Recent grants

• AF: Medium: Behavioral design for online environments

  NSF · $1.2M · 2015–2021

• Combinatorial and Algorithmic Aspects of Network Coding

  NSF · $250k · 2007–2011

• PostDoctoral Research Fellowship

  NSF · $108k · 2005–2009

• CAREER: Algorithms for Environments with Incomplete Information

  NSF · $400k · 2007–2013

Frequent coauthors

• Maria-Florina Balcan

  121 shared

• Yuval Rabani

  121 shared

• Michael Dinitz

  121 shared

• Shubhangi Saraf

  Rutgers, The State University of New Jersey

  121 shared

• Google Safetoc

  Hebrew University of Jerusalem

  121 shared

• Brock R. Baker

  121 shared

• Sandy Irani

  121 shared

• Avrim Blum

  121 shared

Labs

• Robert Kleinberg's LabPI

  Focuses on algorithms and theoretical computer science, especially economic aspects of algorithms, online learning and its applications, and random processes in networks.

Education

• Ph.D.

  Massachusetts Institute of Technology

  2005

Awards & honors

• Microsoft Research New Faculty Fellowship
• Alfred P. Sloan Foundation Fellowship
• NSF CAREER Award
• ACM Fellow (2021)