Shie Mannor

extended abstract Calibrated Forecasts: Efficiency versus Universality (2008)

Kernel methods have become popular in many sub-fields of machine learning, with the exception of reinforcement learning; they facilitate rich representations, and enable machine learning techniques...

Gürdal Arslan, Shie Mannor, Jeff S. Shamma

One approach to learning/adaptation in repeated matrix games is to have each player compute some sort of forecast of opponent actions and play a best response to this forecast. Accordingly, the...

An Inequality for Nearly Log-Concave Distributions With Applications to Learning (2008)

Constantine Caramanis, Shie Mannor

Abstract—We prove that given a nearly log-concave distribution, in any partition of the space to two well separated sets, the measure of the points that do not belong to these sets is large. We...

Strategies for prediction under imperfect monitoring (2008)

Gábor Lugosi, Shie Mannor, Gilles Stoltz

3 CNRS and Département de mathématiques et applications, Ecole normale

Reinforcement Learning-based Load Shared Sequential Routing (2008)

Fariba Heidari, Shie Mannor, Lorne G. Mason

Abstract. We consider event dependent routing algorithms for on-line explicit source routing in MPLS networks. The proposed methods are based on load shared sequential routing in which load sharing...

Robustness and Regularization of Support Vector Machines (2008)

Xu, Huan, Caramanis, Constantine, Mannor, Shie

We consider regularized support vector machines (SVMs) and show that they are precisely equivalent to a new robust optimization formulation. We show that this equivalence of robust optimization and...

Shie Mannor, Duncan Simester, Peng Sun, John N. Tsitsiklis

informs ® doi 10.1287/mnsc.1060.0614

The Robustness-Performance Tradeoff in Markov Decision Processes (2008)

Huan Xu, Shie Mannor

Computation of a satisfactory control policy for a Markov decision process when the parameters of the model are not exactly known is a problem encountered in many practical applications. The...

Percentile Optimization in Uncertain Markov Decision Processes with Application to Efficient Exploration (2008)

Erick Delage, Shie Mannor

Markov decision processes are an effective tool in modeling decision-making in uncertain dynamic environments. Since the parameters of these models are typically estimated from data, learned from...

Strategies for prediction under imperfect monitoring (2008)

Gábor Lugosi, Shie Mannor, Gilles Stoltz

3 CNRS and Département de mathématiques et applications, Ecole normale

Strategies for prediction under imperfect monitoring (2008)

Gábor Lugosi, Shie Mannor, Gilles Stoltz

We propose simple randomized strategies for sequential decision (or prediction) under imperfect monitoring, that is, when the decision maker (forecaster) does not have access to the past outcomes but...

Approachability in Repeated Games: Computational Aspects and a Stackelberg Variant 1 (2008)

Strategies for prediction under imperfect monitoring (2008)

We consider a finite two-player zero-sum game with vector-valued rewards. We study the question of whether a given polyhedral set D is “approachable, ” that is, whether Player 1 (the “decision...

Strategies for prediction under imperfect monitoring (2008)

We propose simple randomized strategies for sequential prediction under imperfect monitoring, that is, when the forecaster does not have access to the past outcomes but rather to a feedback signal....

Strategies for prediction under imperfect monitoring. (2008)

We propose simple randomized strategies for sequential prediction under imperfect monitoring, that is, when the forecaster does not have access to the past outcomes but rather to a feedback signal....

Lugosi, Gábor, Mannor, Shie, Stoltz, Gilles

We propose simple randomized strategies for sequential decision (or prediction) under imperfect monitoring, that is, when the decision maker (forecaster) does not have access to the past outcomes but...

Error Reducing Sampling in Reinforcement Learning (2008)

Error Reducing Sampling in Reinforcement Learning (2008)

In reinforcement learning, an agent collects information interacting with an environment and uses it to derive a behavior. This paper focuses on efficient sampling; that is, the problem of choosing...

Regularized Fitted Q-iteration: Application to Planning (2008)

In reinforcement learning, an agent collects information interacting with an environment and uses it to derive a behavior. This paper focuses on efficient sampling; that is, the problem of choosing...

Farahmand, Amir Massoud, Ghavamzadeh, Mohammad, Mannor, Shie, Szepesvari, Csaba

We consider planning in a Markovian decision problem, i.e., the problem of finding a good policy given access to a generative model of the environment. We propose to use fitted Q-iteration with...

Large Deviations Bounds for Markov Decision Processes Under General Policies (2007)

The Empirical Bayes Envelope and Regret Minimization in Stochastic Games (2007)

We consider the empirical state-action frequencies and the empirical reward in weakly communicating finite-state Markov decision processes under general policies. We define a certain polytope and...

On Universal Compression of Multi-Dimensional Data Arrays Using Self-Similar Curves (2007)

Abstract. In repeated matrix games, classical results establish the existence of regret minimizing strategies that secure an average payo as high as the best-response payo (the Bayes Envelope)...

Tsachy Weissman, Shie Mannor

The LZ-based scheme for the lossless compression of multi-dimensional data arrays using self-similar curves is revisited. The asymptotic universality of this compression scheme is established for the...

Abstract (2007)

Shie Mannor, Ron Meir, Shahar Mendelson

Boosting algorithms have been shown to perform well on many realworld problems, although they sometimes tend to overfit in noisy situations. While excellent finite sample bounds are known, it has not...

Reinforcement Learning for Average Reward Zero-Sum Games (2007)

Zero-sum Games, Shie Mannor

We consider Reinforcement Learning for average reward zerosum stochastic games. We present and analyze two algorithms. The first is based on relative Q-learning and the second on Q-learning for...

Error reducing sampling in reinforcement learning (2007)

Bruno Scherrer, Shie Mannor

In reinforcement learning, an agent collects information interacting with an environment and uses it to derive a behavior. This paper focuses on efficient sampling; that is, the problem of choosing...

Strategies for prediction under imperfect monitoring (2007)

Strategies for prediction under imperfect monitoring (2007)

We propose simple randomized strategies for sequential prediction under imperfect monitoring, that is, when the forecaster does not have access to the past outcomes but rather to a feedback signal....

Lugosi, Gábor, Stoltz, Gilles, Mannor, Shie

We propose simple randomized strategies for sequential prediction under imperfect monitoring, that is, when the forecaster does not have access to the past outcomes but rather to a feedback signal....

Strategies for prediction under imperfect monitoring (2007)

Strategies for prediction under imperfect monitoring (2007)

We propose simple randomized strategies for sequential prediction under imperfect monitoring, that is, when the forecaster does not have access to the past outcomes but rather to a feedback signal....

Strategies for prediction under imperfect monitoring (2007)

We propose simple randomized strategies for sequential prediction under imperfect monitoring, that is, when the forecaster does not have access to the past outcomes but rather to a feedback signal....

Strategies for prediction under imperfect monitoring (2007)

We propose simple randomized strategies for sequential prediction under imperfect monitoring, that is, when the forecaster does not have access to the past outcomes but rather to a feedback signal....

Lugosi, Gábor, Stoltz, Gilles, Mannor, Shie

We propose simple randomized strategies for sequential prediction under imperfect monitoring, that is, when the forecaster does not have access to the past outcomes but rather to a feedback signal....

Network Formation: Bilateral Contracting and Myopic Dynamics (2007)

Esteban Arcaute, Ramesh Johari, Shie Mannor

We consider a network formation game where nodes wish to send traffic to each other. Nodes contract bilaterally with each other to form bidirectional communication links; once the network is formed,...

Multi-agent learning for engineers (2007)

Shie Mannor, Jeff S. Shamma

As suggested by the title of Shoham, Powers, and Grenager’s position paper [34], the ultimate lens through which the multi-agent learning framework should be assessed is “what is the...

Adaptive timeout policies for fast finegrained power management (2007)

Branislav Kveton, Shie Mannor

Power management techniques for mobile appliances put the components of the systems into low power states to maximize battery life while minimizing the impact on the perceived performance of the...

The cross-entropy method for power system combinatorial optimization problems (2007)

Ernst, Damien, Glavic, Mevludin, Stan, Guy-Bart, Mannor, Shie, Wehenkel, Louis

2007 Power Tech

Approachability in Repeated Games: Computational Aspects and a Stackelberg Variant 1 (2007)

A contract-based model for directed network formation (2006)

We consider a finite two-player zero-sum game with vector-valued rewards. We study the question of whether a given polyhedral set D is “approachable, ” that is, whether Player 1 (the “decision...

Geometric Variance Reduction in Markov Chains: Application to Value Function and Gradient Estimation (2006)

We consider a network game where the nodes of the network wish to form a graph to route traffic between themselves. We present a model where costs are incurred for routing traffic, as well as for a...

Remi Munos, Shie Mannor

We study a variance reduction technique for Monte Carlo estimation of functionals in Markov chains. The method is based on designing sequential control variates using successive approximations of the...

A Simulation-Based Algorithm for Ergodic Control of Markov Chains Conditioned on Rare Events (2006)

Shalabh Bhatnagar, Vivek S. Borkar, Madhukar Akarapu, Shie Mannor

We study the problem of long-run average cost control of Markov chains conditioned on a rare event. In a related recent work, a simulation based algorithm for estimating performance measures...

Journal of Machine Learning Research 7 (2006) 1079--1105 Submitted 2/05; Published 6/06 Action Elimination and Stopping Conditions for the (2006)

Eyal Even-dar, Shie Mannor, Yishay Mansour, Sridhar Mahadevan

We incorporate statistical confidence intervals in both the multi-armed bandit and the reinforcement learning problems. In the bandit problem we show that given n arms, it suffices to pull the arms a...

Evolutionary function approximation for reinforcement learning (2006)

Shimon Whiteson, Peter Stone, Sven Koenig, Shie Mannor, Georgios Theocharous

Temporal difference methods are theoretically grounded and empirically effective methods for addressing reinforcement learning problems. In most real-world reinforcement learning tasks, TD methods...

Automatic basis function construction for approximate dynamic programming and reinforcement learning (2006)

Shie Mannor, Doina Precup

We address the problem of automatically constructing basis functions for linear approximation of the value function of a Markov Decision Process (MDP). Our work builds on results by Bertsekas and...

Evolutionary function approximation for reinforcement learning (2006)

Shimon Whiteson, Peter Stone, Sven Koenig, Shie Mannor, Georgios Theocharous

Temporal difference methods are theoretically grounded and empirically effective methods for addressing reinforcement learning problems. In most real-world reinforcement learning tasks, TD methods...

Asymptotics of Efficiency Loss in Competitive Market Mechanisms (2006)

Jia Yuan Yu, Shie Mannor

• Problem: allocation of bandwidth in a network. – Feature: heterogeneous demand.

Reinforcement learning with Gaussian processes (2005)

Engel, Yaakov, Mannor, Shie, Meir, Ron

Gaussian Process Temporal Difference (GPTD) learning offers a Bayesian solution to the policy evaluation problem of reinforcement learning. In this paper we extend the GPTD framework by addressing...

Efficiency Loss in a Network Resource Allocation Game: The Case of Elastic Supply (2005)

Johari, Ramesh, Mannor, Shie, Tsitsiklis, John N.

We consider a resource allocation problem where individual users wish to send data across a network to maximize their utility, and a cost is incurred at each link that depends on the total rate sent...

Basis function adaptation in temporal difference reinforcement learning (2005)

Basis function adaptation in temporal difference reinforcement learning (2005)

Reinforcement Learning (RL) is an approach for solving complex multistage decision problems that fall under the general framework of Markov Decision Problems (MDPs), with possibly unknown parameters....

Reinforcement learning with Gaussian processes (2005)

∗ All correspondence should be sent to this author. 1 We examine methods for on-line optimization of the basis function for temporal difference Reinforcement Learning algorithms. We concentrate on...

Reinforcement learning with Gaussian processes (2005)

Gaussian Process Temporal Difference (GPTD) learning offers a Bayesian solution to the policy evaluation problem of reinforcement learning. In this paper we extend the GPTD framework by addressing...

Reinforcement learning with Gaussian processes (2005)

Gaussian Process Temporal Difference (GPTD) learning offers a Bayesian solution to the policy evaluation problem of reinforcement learning. In this paper we extend the GPTD framework by addressing...

Approachability in Repeated Games: Computational Aspects and a Stackelberg Variant 1 (2005)

Gaussian Process Temporal Difference (GPTD) learning offers a Bayesian solution to the policy evaluation problem of reinforcement learning. In this paper we extend the GPTD framework by addressing...

We consider a finite two-player zero-sum game with vector-valued rewards. We study the question of whether a given polyhedral set D is “approachable, ” that is, whether Player 1 (the “decision...

De Boer, Pieter-Tjerk, Kroese, D. P., Mannor, Shie, Rubinstein, R. Y.

The cross-entropy approach is a new generic approach to combinatorial and multi-extremal optimization and rare event simulation. The purpose of this tutorial is to give a gentle introduction to the...

De Boer, Pieter-Tjerk, Kroese, D. P., Mannor, Shie, Rubinstein, R. Y.

The cross-entropy approach is a new generic approach to combinatorial and multi-extremal optimization and rare event simulation. The purpose of this tutorial is to give a gentle introduction to the...

The Kernel recursive least squares algorithm (2004)

Engel, Yaakov, Mannor, Shie, Meir, Ron

We present a non-linear kernel-based version of the Recursive Least Squares (RLS) algorithm. Our Kernel-RLS (KRLS) algorithm performs linear regression in the feature space induced by a Mercer...

Error reducing sampling in reinforcement learning (2004)

Error reducing sampling in reinforcement learning (2004)

In reinforcement learning, an agent collects information interacting with an environment and uses it to derive a behavior. This paper focuses on efficient sampling; that is, the problem of choosing...

Error reducing sampling in reinforcement learning (2004)

In reinforcement learning, an agent collects information interacting with an environment and uses it to derive a behavior. This paper focuses on efficient sampling; that is, the problem of choosing...

In reinforcement learning, an agent collects information interacting with an environment and uses it to derive a behavior. This paper focuses on efficient sampling; that is, the problem of choosing...

De Boer, Pieter-Tjerk, Kroese, D. P., Mannor, Shie, Rubinstein, R. Y.

The cross-entropy approach is a new generic approach to combinatorial and multi-extremal optimization and rare event simulation. The purpose of this tutorial is to give a gentle introduction to the...

De Boer, Pieter-Tjerk, Kroese, D. P., Mannor, Shie, Rubinstein, R. Y.

The cross-entropy approach is a new generic approach to combinatorial and multi-extremal optimization and rare event simulation. The purpose of this tutorial is to give a gentle introduction to the...

The sample complexity of exploration in the multi-armed bandit problem (2004)

Shie Mannor, John N. Tsitsiklis, Kristin Bennett, Nicolò Cesa-bianchi

We consider the multi-armed bandit problem under the PAC (“probably approximately correct”) model. It was shown by Even-Dar et al. (2002) that given n arms, a total of O � (n/ε 2)log(1/δ) �...

Dynamic abstraction in reinforcement learning via clustering (2004)

Shie Mannor, Ishai Menache, Amit Hoze, Uri Klein

We consider a graph theoretic approach for automatic construction of options in a dynamic environment. A map of the environment is generated on-line by the learning agent, representing the...

Kernel recursive least squares (2004)

Bias and variance in value function estimation (2004)

We present a non-linear kernel-based version of the Recursive Least Squares (RLS) algorithm. Our Kernel-RLS algorithm performs linear regression in the feature space induced by a Mercer kernel, and...

Shie Mannor, Peng Sun, John N. Tsitsiklis

We consider the bias and variance of value function estimation that are caused by using an empirical model instead of the true model. We analyze these bias and variance for Markov processes from a...

The Kernel Recursive Least Squares Algorithm (2004)

A geometric approach to multi-criterion reinforcement learning (2004)

Copyright c ○ 2003 by the authors. This report may be used, copied and distributed in its entirety for academic purposes. Any copies must include this cover page and notice. Any non-academic use...

Efficiency Loss in a Network Resource Allocation Game: The Case of Elastic Supply (2004)

We consider the problem of reinforcement learning in a dynamic environment, where the learning objective is dened in terms of multiple reward functions of the average reward type. The environment is...

Efficiency Loss in a Network Resource Allocation Game: The Case of Elastic Supply (2004)

We consider a resource allocation problem where individual users wish to send data across a network to maximize their utility, and a cost is incurred at each link that depends on the total rate sent...

An Inequality for Nearly Log-concave Distributions with Applications to Learning (2004)

We consider a resource allocation problem where individual users wish to send data across a network to maximize their utility, and a cost is incurred at each link that depends on the total rate sent...

Constantine Caramanis, Shie Mannor

We prove that given a nearly log-concave density, in any partition of the space to two well separated sets, the measure of the points that do not belong to these sets is large. We apply this...

On the Empirical State-Action Frequencies in Markov Decision Processes Under General Policies (2004)

Efficiency loss in a network resource allocation game (2004)

We consider the empirical state-action frequencies and the empirical reward in weakly communicating finite-state Markov decision processes under general policies. We define a certain polytope and...

A tutorial on the cross-entropy method (2004)

Abstract—We consider a resource allocation problem where individual users wish to send data across a network to maximize their utility, and a cost is incurred at each link that depends on the total...

Dirk P. Kroese, Shie Mannor, Reuven Y. Rubinstein

The cross-entropy (CE) method is a new generic approach to combinatorial and multi-extremal optimization and rare event simulation. The purpose of this tutorial is to give a gentle introduction to...

A Geometric Approach to Multi-Criterion Reinforcement Learning (2004)

Shie Mannor, Nahum Shimkin, Sridhar Mahadevan

We consider the problem of reinforcement learning in a controlled Markov environment with multiple objective functions of the long-term average reward type. The environment is initially unknown, and...

Bayes meets Bellman: The Gaussian process approach to temporal difference learning (2003)

Greedy algorithms for classification - consistency, convergence rates, and adaptivity (2003)

We present a novel Bayesian approach to the problem of value function estimation in continuous state spaces. We dene a probabilistic generative model for the value function by imposing a Gaussian...

Shie Mannor, Ron Meir, Tong Zhang

Many regression and classification algorithms proposed over the years can described as greedy procedures for the stagewise minimization of an appropriate cost function. Some examples include additive...

Action elimination and stopping conditions for reinforcement learning (2003)

The cross entropy method for fast policy search (2003)

We consider incorporating action elimination procedures in reinforcement learning algorithms. We suggest a framework that is based on learning an upper and a lower estimates of the value function or...

Shie Mannor, Reuven Rubinstein, Yohai Gat

We present a learning framework for Markovian decision processes that is based on optimization in the policy space. Instead of using relatively slow gradient-based optimization algorithms, we use the...

The empirical Bayes envelope and regret minimization in competitive Markov decision processes (2003)

A Contract-Based Model for Directed Network Formation (2003)

Abstract. This paper proposes an extension of the regret minimizing framework from re-peated matrix games to stochastic game models, under appropriate recurrence conditions. A decision maker (P1) who...

Greedy Algorithms for Classification - Consistency, Convergence Rates, and Adaptivity (2003)

We consider a network game where the nodes of the network wish to form a graph to route traffic between themselves. We present a model where costs are incurred for routing traffic, as well as for a...

Shie Mannor, Ron Meir, Tong Zhang, Yoram Singer

Many regression and classification algorithms proposed over the years can be described as greedy procedures for the stagewise minimization of an appropriate cost function. Some examples include...

Action elimination and stopping conditions for reinforcement learning (2003)

We consider incorporating action elimination procedures in reinforcement learning algorithms. We suggest a framework that is based on learning an upper and a lower estimates of the value function or...

Bayes Meets Bellman: The Gaussian Process Approach to Temporal Difference Learning (2003)

doi 10.1287/moor.1050.0148

Lower Bounds on the Sample Complexity of Exploration in the Multi-Armed Bandit Problem (2003)

We present a novel Bayesian approach to the problem of value function estimation in continuous state spaces. We define a probabilistic generative model for the value function by imposing a Gaussian...

On-line Learning with Imperfect Monitoring (2003)

We consider the Multi-armed bandit problem under the PAC ("probably approximately correct") model. It was shown by Even-Dar et al. [5] that given n arms, it su#ces to play the arms a total...

Q-Cut - dynamic discovery of sub-goals in reinforcement learning (2002)

We study on-line play of repeated matrix games in which the observations of past actions of the other player and the obtained reward are partial and stochastic. We define the Partial Observation...

PAC bounds for multi-armed bandit and Markov decision processes (2002)

Abstract. We present the Q-Cut algorithm, a graph theoretic approach for automatic detection of sub-goals in a dynamic environment, which is used for acceleration of the Q-Learning algorithm. The...

Sparse online greedy support vector regression (2002)

Abstract. The bandit problem is revisited and considered under the PAC model. Our main contribution in this part is to show that given n arms, it suffices to pull the arms O ( n ɛ2 log 1) times to...

PAC bounds for multi-armed bandit and Markov decision processes (2002)

Abstract. We present a novel algorithm for sparse online greedy kernelbased nonlinear regression. This algorithm improves current approaches to kernel-based regression in two aspects. First, it...

Sparse online greedy support vector regression (2002)

1 Introduction The Multi-armed bandit problem is one of the classical problems in decisiontheory. The problem is very simple to model- there are a number of alternative arms, each with a stochastic...

The consistency of greedy algorithms for classification (2002)

Abstract. We present a novel algorithm for sparse online greedy kernelbased nonlinear regression. This algorithm improves current approaches to kernel-based regression in two aspects. First, it...

Shie Mannor, Ron Meir, Tong Zhang

Abstract. We consider a class of algorithms for classification, which are based on sequential greedy minimization of a convex upper bound on the 0- 1 loss function. A large class of recently popular...

The consistency of greedy algorithms for classification (2002)

Shie Mannor, Ron Meir, Tong Zhang

Abstract. We consider a class of algorithms for classification, which are based on sequential greedy minimization of a convex upper bound on the 0- 1 loss function. A large class of recently popular...

The Steering Approach for Multi-Criteria Reinforcement (2002)

The Steering Approach for Multi-Criteria Reinforcement (2002)

We consider the problem of learning to attain multiple goals in a dynamic environment, which is initially unknown. In addition, the environment may contain arbitrarily varying elements related to...

Sparse Online Greedy Support Vector Regression (2002)

We consider the problem of learning to attain multiple goals in a dynamic environment, which is initially unknown. In addition, the environment may contain arbitrarily varying elements related to...

On the Existence of Linear Weak Learners and Applications to Boosting (2002)

We present a novel algorithm for sparse online greedy kernelbased nonlinear regression. This algorithm improves current approaches to kernel-based regression in two aspects. First, it operates online...

Shie Mannor, Ron Meir, Yoshua Bengio, Dale Schuurmans

We consider the existence of a linear weak learner for boosting algorithms. A weak learner for binary classification problems is required to achieve a weighted empirical error on the training set...

Sparse Online Greedy Support Vector Regression (2002)

The Steering Approach for Multi-Criteria Reinforcement (2002)

We present a novel algorithm for sparse online greedy kernelbased nonlinear regression. This algorithm improves current approaches to kernel-based regression in two aspects. First, it operates online...

Q-Cut - Dynamic Discovery of Sub-Goals in Reinforcement Learning (2002)

We consider the problem of learning to attain multiple goals in a dynamic environment, which is initially unknown. In addition, the environment may contain arbitrarily varying elements related to...

Adaptive strategies and regret minimization in arbitrarily varying Markov environments (2001)

We present the Q-Cut algorithm, a graph theoretic approach for automatic detection of sub-goals in a dynamic environment, which is used for acceleration of the Q-Learning algorithm. The learning...

Geometric bounds for generalization in boosting (2001)

Abstract. We consider the problem of maximizing the average reward in a controlled Markov environment, which also contains some arbitrarily varying elements. This problem is captured by a two-person...

Weak learners and improved rates of convergence in boosting (2001)

Abstract. We consider geometric conditions on a labeled data set which guarantee that boosting algorithms work well when linear classiers are used as weak learners. We start by providing conditions...

On Finding Good State Aggregation Functions (2001)

The problem constructing weak classifiers for boosting algorithms is studied. We present an algorithm that produces a linear classifier that is guaranteed to achieve an error better than random...

Yaakov Engel, Shie Mannor

We describe a novel algorithm that learns to

Learning embedded maps of Markov processes (2001)

Yaakov Engel, Shie Mannor

We present HEMP- a novel learning algorithm designed to operate in the domain of Markov processes and Markov decision processes. HEMP learns to perform a Heuristic Embedding of Markov Processes into...

Weak learners and improved rates of convergence in boosting (2001)

Xilinx Inc.: XPower Tutorial: FPGA Design, XPower (2001)

The problem of constructing weak classiers for boosting algorithms is studied. We present an algorithm that produces a linear classi er that is guaranteed to achieve an error better than random...

Geometric bounds for generalization in boosting (2001)

Abstract. We study online learning where the objective of the decision maker is to maximize her average long-term reward given that some average constraints are satisfied along the sample path. We...

Xilinx Inc.: XPower Tutorial: FPGA Design, XPower (2001)

Abstract. We consider geometric conditions on a labeled data set which guarantee that boosting algorithms work well when linear classifiers are used as weak learners. We start by providing conditions...

Regret minimization in signal space for repeated matrix games with partial observations (2000)

Abstract. We study online learning where the objective of the decision maker is to maximize her average long-term reward given that some average constraints are satisfied along the sample path. We...

Regret minimization in signal space for repeated matrix games with partial observations (2000)

Abstract. The Bayes Envelope of a repeated matrix game traces the maximal payo rate that a player (say P1) could secure for herself had she known in advance the empirical frequency of P2's...

Generalized approachability results for stochastic games with a single communicating state (2000)

Abstract. The Bayes Envelope of a repeated matrix game traces the maximal payoff rate that a player (say P1) could secure for herself had she known in advance the empirical frequency of P2's...