We address the problem of executing large client orders in continuous double-auction markets under time and liquidity constraints. We propose a model predictive control (MPC) framework that balances three competing objectives: order completion, market impact, and opportunity cost. Our algorithm is guided by a trading schedule (such as time-weighted average price or volume-weighted average price) but allows for deviations to reduce the expected execution cost, with due regard to risk. Our MPC algorithm executes the order progressively, and at each decision step it solves a fast quadratic program that trades off expected transaction cost against schedule deviation, while incorporating a residual cost term derived from a simple base policy. Approximate schedule adherence is maintained through explicit bounds, while variance constraints on deviation provide direct risk control. The resulting system is modular, data-driven, and suitable for deployment in production trading infrastructure. Using six months of NASDAQ 'level 3' data and simulated orders, we show that our MPC approach reduces schedule shortfall by approximately 40-50

Evolving security vulnerabilities and shifting operational conditions require frequent updates to network security policies. These updates include adjustments to incident response procedures and modifications to access controls, among others. Reinforcement learning methods have been proposed for automating such policy adaptations, but most of the methods in the research literature lack performance guarantees and adapt slowly to changes. In this paper, we address these limitations and present a method for computing security policies that is scalable, offers theoretical guarantees, and adapts quickly to changes. It assumes a model or simulator of the system and comprises three components: belief estimation through particle filtering, offline policy computation through aggregation, and online policy adaptation through rollout. Central to our method is a new feature-based aggregation technique, which improves scalability and flexibility. We analyze the approximation error of aggregation and show that rollout efficiently adapts policies to changes under certain conditions. Simulations and testbed results demonstrate that our method outperforms state-of-the-art methods on several benchmarks, including CAGE-2.

We consider deterministic finite-horizon optimal control problems with a fixed initial state. We introduce an on-line policy iteration method, which starting from a given policy, however obtained, generates a sequence of cost improving policies and corresponding trajectories. Each policy produces a trajectory, which is used in turn to generate data for training the next policy. The method is motivated by problems that are repeatedly solved starting from the same initial state, including discrete optimization and path planning for repetitive tasks. For such problems, the method is fast enough to be used on-line. Under a natural consistency condition, we show that the sequence of costs of the generated policies is monotonically improving for the given initial state (but not necessarily for other states). We illustrate our results with computational studies from combinatorial optimization and 3-dimensional path planning for drones in the presence of obstacles. We also discuss briefly a stochastic counterpart of our algorithm. Our proposed framework combines elements of rollout and policy iteration with flexible trajectory-based policy representations, and applies to problems involving a single as well as multiple decision makers. It also provides a principled way to train neural network-based policies using trajectory data, while preserving monotonic cost improvement.

Presents reviews for the following list of books, (Rollout, Policy Iteration, and Distributed Reinforcement Learning).

In this work we consider a generalization of the well-known multivehicle routing problem: given a network, a set of agents occupying a subset of its nodes, and a set of tasks, we seek a minimum cost sequence of movements subject to the constraint that each task is visited by some agent at least once. The classical version of this problem assumes a central computational server that observes the entire state of the system perfectly and directs individual agents according to a centralized control scheme. In contrast, we assume that there is no centralized server and that each agent is an individual processor with no a priori knowledge of the underlying network (including task and agent locations). Moreover, our agents possess strictly local communication and sensing capabilities (restricted to a fixed radius around their respective locations), aligning more closely with several real-world multiagent applications. These restrictions introduce many challenges that are overcome through local information sharing and direct coordination between agents. We present a fully distributed, online, and scalable reinforcement learning algorithm for this problem whereby agents self-organize into local clusters and independently apply a multiagent rollout scheme locally to each cluster. We demonstrate empirically via extensive simulations that there exists a critical sensing radius beyond which the distributed rollout algorithm begins to improve over a greedy base policy. This critical sensing radius grows proportionally to the $\log^*$ function of the size of the network, and is, therefore, a small constant for any relevant network. Our decentralized reinforcement learning algorithm achieves approximately a factor of two cost improvement over the base policy for a range of radii bounded from below and above by two and three times the critical sensing radius, respectively.

In this paper we consider a transformer with an n-gram structure, such as the one underlying ChatGPT. The transformer provides next word probabilities, which can be used to generate word sequences. We consider methods for computing word sequences that are highly likely, based on these probabilities. Computing the optimal (i.e., most likely) word sequence starting with a given initial state is an intractable problem, so we propose methods to compute highly likely sequences of N words in time that is a low order polynomial in N and in the vocabulary size of the n-gram. These methods are based on the rollout approach from approximate dynamic programming, a form of single policy iteration, which can improve the performance of any given heuristic policy. In our case we use a greedy heuristic that generates as next word one that has the highest probability. We show with analysis, examples, and computational experimentation that our methods are capable of generating highly likely sequences with a modest increase in computation over the greedy heuristic. While our analysis and experiments are focused on Markov chains of the type arising in transformer and ChatGPT-like models, our methods apply to general finite-state Markov chains, and related inference applications of Hidden Markov Models (HMM), where Viterbi decoding is used extensively.

In this paper we apply model predictive control (MPC), rollout, and reinforcement learning (RL) methodologies to computer chess. We introduce a new architecture for move selection, within which available chess engines are used as components. One engine is used to provide position evaluations in an approximation in value space MPC/RL scheme, while a second engine is used as nominal opponent, to emulate or approximate the moves of the true opponent player. We show that our architecture improves substantially the performance of the position evaluation engine. In other words our architecture provides an additional layer of intelligence, on top of the intelligence of the engines on which it is based. This is true for any engine, regardless of its strength: top engines such as Stockfish and Komodo Dragon (of varying strengths), as well as weaker engines. Structurally, our basic architecture selects moves by a one-move lookahead search, with an intermediate move generated by a nominal opponent engine, and followed by a position evaluation by another chess engine. Simpler schemes that forego the use of the nominal opponent, also perform better than the position evaluator, but not quite by as much. More complex schemes, involving multistep lookahead, may also be used and generally tend to perform better as the length of the lookahead increases. Theoretically, our methodology relies on generic cost improvement properties and the superlinear convergence framework of Newton's method, which fundamentally underlies approximation in value space, and related MPC/RL and rollout/policy iteration schemes. A critical requirement of this framework is that the first lookahead step should be executed exactly. This fact has guided our architectural choices, and is apparently an important factor in improving the performance of even the best available chess engines.

In this work, we consider data association problems involving multi-object tracking (MOT). In particular, we address the challenges arising from object occlusions. We propose a framework called approximate dynamic programming track (ADPTrack), which applies dynamic programming principles to improve an existing method called the base heuristic. Given a set of tracks and the next target frame, the base heuristic extends the tracks by matching them to the objects of this target frame directly. In contrast, ADPTrack first processes a few subsequent frames and applies the base heuristic starting from the next target frame to obtain tentative tracks. It then leverages the tentative tracks to match the objects of the target frame. This tends to reduce the occlusion-based errors and leads to an improvement over the base heuristic. When tested on the MOT17 video dataset, the proposed method demonstrates a 0.7 association accuracy (IDF1 metric) over a state-of-the-art method that is used as the base heuristic. It also obtains improvements with respect to all the other standard metrics. Empirically, we found that the improvements are particularly pronounced in scenarios where the video data is obtained by fixed-position cameras.

In this paper, we focus on the autonomous multiagent taxi routing problem for a large urban environment where the location and number of future ride requests are unknown a-priori, but can be estimated by an empirical distribution. Recent theory has shown that a rollout algorithm with a stable base policy produces a near-optimal stable policy. In the routing setting, a policy is stable if its execution keeps the number of outstanding requests uniformly bounded over time. Although, rollout-based approaches are well-suited for learning cooperative multiagent policies with considerations for future demand, applying such methods to a large urban environment can be computationally expensive due to the large number of taxis required for stability. In this paper, we aim to address the computational bottleneck of multiagent rollout by proposing an approximate multiagent rollout-based two phase algorithm that reduces computational costs, while still achieving a stable near-optimal policy. Our approach partitions the graph into sectors based on the predicted demand and the maximum number of taxis that can run sequentially given the user's computational resources. The algorithm then applies instantaneous assignment (IA) for re-balancing taxis across sectors and a sector-wide multiagent rollout algorithm that is executed in parallel for each sector. We provide two main theoretical results: 1) characterize the number of taxis $m$ that is sufficient for IA to be stable; 2) derive a necessary condition on $m$ to maintain stability for IA as time goes to infinity. Our numerical results show that our approach achieves stability for an $m$ that satisfies the theoretical conditions. We also empirically demonstrate that our proposed two phase algorithm has equivalent performance to the one-at-a-time rollout over the entire map, but with significantly lower runtimes.

We consider the classical linear assignment problem, and we introduce new auction algorithms for its optimal and suboptimal solution. The algorithms are founded on duality theory, and are related to ideas of competitive bidding by persons for objects and the attendant market equilibrium, which underlie real-life auction processes. We distinguish between two fundamentally different types of bidding mechanisms: aggressive and cooperative. Mathematically, aggressive bidding relies on a notion of approximate coordinate descent in dual space, an epsilon-complementary slackness condition to regulate the amount of descent approximation, and the idea of epsilon-scaling to resolve efficiently the price wars that occur naturally as multiple bidders compete for a smaller number of valuable objects. Cooperative bidding avoids price wars through detection and cooperative resolution of any competitive impasse that involves a group of persons. We discuss the relations between the aggressive and the cooperative bidding approaches, we derive new algorithms and variations that combine ideas from both of them, and we also make connections with other primal-dual methods, including the Hungarian method. Furthermore, our discussion points the way to algorithmic extensions that apply more broadly to network optimization, including shortest path, max-flow, transportation, and minimum cost flow problems with both linear and convex cost functions.

In this paper we describe a new conceptual framework that connects approximate Dynamic Programming (DP), Model Predictive Control (MPC), and Reinforcement Learning (RL). This framework centers around two algorithms, which are designed largely independently of each other and operate in synergy through the powerful mechanism of Newton's method. We call them the off-line training and the on-line play algorithms. The names are borrowed from some of the major successes of RL involving games; primary examples are the recent (2017) AlphaZero program (which plays chess, [SHS17], [SSS17]), and the similarly structured and earlier (1990s) TD-Gammon program (which plays backgammon, [Tes94], [Tes95], [TeG96]). In these game contexts, the off-line training algorithm is the method used to teach the program how to evaluate positions and to generate good moves at any given position, while the on-line play algorithm is the method used to play in real time against human or computer opponents. Significantly, the synergy between off-line training and on-line play also underlies MPC (as well as other major classes of sequential decision problems), and indeed the MPC design architecture is very similar to the one of AlphaZero and TD-Gammon. This conceptual insight provides a vehicle for bridging the cultural gap between RL and MPC, and sheds new light on some fundamental issues in MPC. These include the enhancement of stability properties through rollout, the treatment of uncertainty through the use of certainty equivalence, the resilience of MPC in adaptive control settings that involve changing system parameters, and the insights provided by the superlinear performance bounds implied by Newton's method.

In this paper, we consider an important class of Partially Observable Markov Decision Processes (POMDP) with unknown parameters, which contains the Wordle puzzle as a special case. For this class of POMDP, we develop a new on-line solution method, which is based on the rollout approach. Our method relies on the use of a base heuristic policy and guarantees cost improvement over that policy. When applied to Wordle, our algorithm solves the puzzle on-line. The performance is within 0.4% of the known optimal results, and is substantially better than that of the base heuristic policies we have tested.

We derive a learning framework to generate routing/pickup policies for a fleet of autonomous vehicles tasked with servicing stochastically appearing requests on a city map. We focus on policies that 1) give rise to coordination amongst the vehicles, thereby reducing wait times for servicing requests, 2) are non-myopic, and consider a-priori potential future requests, 3) can adapt to changes in the underlying demand distribution. Specifically, we are interested in policies that are adaptive to fluctuations of actual demand conditions in urban environments, such as on-peak vs. off-peak hours. We achieve this through a combination of (i) an online play algorithm that improves the performance of an offline-trained policy, and (ii) an offline approximation scheme that allows for adapting to changes in the underlying demand model. In particular, we achieve adaptivity of our learned policy to different demand distributions by quantifying a region of validity using the q-valid radius of a Wasserstein Ambiguity Set. We propose a mechanism for switching the originally trained offline approximation when the current demand is outside the original validity region. In this case, we propose to use an offline architecture, trained on a historical demand model that is closer to the current demand in terms of Wasserstein distance. We learn routing and pickup policies over real taxicab requests in San Francisco with high variability between on-peak and off-peak hours, demonstrating the ability of our method to adapt to real fluctuation in demand distributions. Our numerical results demonstrate that our method outperforms alternative rollout-based reinforcement learning schemes, as well as other classical methods from operations research.

Knowledge graphs are graph-based data models which can represent real-time data that is constantly growing with the addition of new information. The question-answering systems over knowledge graphs (KGQA) retrieve answers to a natural language question from the knowledge graph. Most existing KGQA systems use static knowledge bases for offline training. After deployment, they fail to learn from unseen new entities added to the graph. There is a need for dynamic algorithms which can adapt to the evolving graphs and give interpretable results. In this research work, we propose using new auction algorithms for question answering over knowledge graphs. These algorithms can adapt to changing environments in real-time, making them suitable for offline and online training. An auction algorithm computes paths connecting an origin node to one or more destination nodes in a directed graph and uses node prices to guide the search for the path. The prices are initially assigned arbitrarily and updated dynamically based on defined rules. The algorithm navigates the graph from the high-price to the low-price nodes. When new nodes and edges are dynamically added or removed in an evolving knowledge graph, the algorithm can adapt by reusing the prices of existing nodes and assigning arbitrary prices to the new nodes. For subsequent related searches, the "learned" prices provide the means to "transfer knowledge" and act as a "guide": to steer it toward the lower-priced nodes. Our approach reduces the search computational effort by 60% in our experiments, thus making the algorithm computationally efficient. The resulting path given by the algorithm can be mapped to the attributes of entities and relations in knowledge graphs to provide an explainable answer to the query. We discuss some applications for which our method can be used.

This paper describes a new algorithm for solving the classical assignment problem. The algorithm is of a primal-dual nature and in some ways resembles the Hungarian and subgradient methods, but is substantially different in other respects. Its main feature is that it is well suited for distributed operation whereby each node participates in the computation on the basis of limited local information about the topology of the network and the data of the problem. The algorithmic process resembles an auction where economic agents compete for resources by making successively higher bids. The algorithm terminates in a finite number of iterations after resource prices reach levels where no further bidding is profitable.

We provide a unifying approximate dynamic programming framework that applies to a broad variety of problems involving sequential estimation. We consider first the construction of surrogate cost functions for the purposes of optimization, and we focus on the special case of Bayesian optimization, using the rollout algorithm and some of its variations. We then discuss the more general case of sequential estimation of a random vector using optimal measurement selection, and its application to problems of stochastic and adaptive control. We distinguish between adaptive control of deterministic and stochastic systems: the former are better suited for the use of rollout, while the latter are well suited for the use of rollout with certainty equivalence approximations. As an example of the deterministic case, we discuss sequential decoding problems, and a rollout algorithm for the approximate solution of the Wordle and Mastermind puzzles, recently developed in the paper [BBB22].

The purpose of this paper is to propose and develop a new conceptual framework for approximate Dynamic Programming (DP) and Reinforcement Learning (RL). This framework centers around two algorithms, which are designed largely independently of each other and operate in synergy through the powerful mechanism of Newton’s method. We call these the off-line training and the on-line play algorithms; the names are borrowed from some of the major successes of RL involving games. Primary examples are the recent (2017) AlphaZero program (which plays chess), and the similarly structured and earlier (1990s) TD-Gammon program (which plays backgammon). In these game contexts, the off-line training algorithm is the method used to teach the program how to evaluate positions and to generate good moves at any given position, while the on-line play algorithm is the method used to play in real time against human or computer opponents.Both AlphaZero and TD-Gammon were trained off-line extensively using neural networks and an approximate version of the fundamental DP algorithm of policy iteration. Yet the AlphaZero player that was obtained off-line is not used directly during on-line play (it is too inaccurate due to approximation errors that are inherent in off-line neural network training). Instead a separate on-line player is used to select moves, based on multistep lookahead minimization and a terminal position evaluator that was trained using experience with the off-line player. The on-line player performs a form of policy improvement, which is not degraded by neural network approximations. As a result, it greatly improves the performance of the off-line player.Similarly, TD-Gammon performs on-line a policy improvement step using one-step or two-step lookahead minimization, which is not degraded by neural network approximations. To this end it uses an off-line neural network-trained terminal position evaluator, and importantly it also extends its on-line lookahead by rollout (simulation with the one-step lookahead player that is based on the position evaluator).An important lesson from AlphaZero and TD-Gammon is that the performance of an off-line trained policy can be greatly improved by on-line approximation in value space, with long lookahead (involving minimization or rollout with the off-line policy, or both), and terminal cost approximation that is obtained off-line. This performance enhancement is often dramatic and is due to a simple fact, which is couched on algorithmic mathematics and is the focal point of this work:(a) Approximation in value space with one-step lookahead minimization amounts to a step of Newton’s method for solving Bellman’s equation.(b) The starting point for the Newton step is based on the results of off-line training, and may be enhanced by longer lookahead minimization and on-line rollout. Indeed the major determinant of the quality of the on-line policy is the Newton step that is performed on-line, while off-line training plays a secondary role by comparison.Significantly, the synergy between off-line training and on-line play also underlies Model Predictive Control (MPC), a major control system design methodology that has been extensively developed since the 1980s. This synergy can be understood in terms of abstract models of infinite horizon DP and simple geometrical constructions, and helps to explain the all-important stability issues within the MPC context. In this work we aim to provide insights (often based on visualization), which explain the beneficial effects of on-line decision making on top of off-line training. In the process, we will bring out the strong connections between the artificial intelligence view of RL, and the control theory views of MPC and adaptive control. While we will deemphasize mathematical proofs, there is considerable related analysis, which supports our conclusions and can be found in the author’s recent RL books (Bertsekas, 2019; Bertsekas, 2020), and the abstract DP monograph (Bertsekas, 2022).One of our principal aims is to show, through the algorithmic ideas of Newton’s method and the unifying principles of abstract DP, that the AlphaZero/TD-Gammon methodology of approximation in value space and rollout applies very broadly to deterministic and stochastic optimal control problems, involving both discrete and continuous search spaces, as well as finite and infinite horizon.

We consider some classical optimization problems in path planning and network transport, and we introduce new auction-based algorithms for their optimal and suboptimal solution. The algorithms are based on mathematical ideas that are related to competitive bidding by persons for objects and the attendant market equilibrium, which underlie auction processes. However, the starting point of our algorithms is different, namely weighted and unweighted path construction in directed graphs, rather than assignment of persons to objects. The new algorithms have several potential advantages over existing methods: they are empirically faster in some important contexts, such as max-flow, they are well-suited for on-line replanning, and they can be adapted to distributed asynchronous operation. Moreover, they allow arbitrary initial prices, without complementary slackness restrictions, and thus are better-suited to take advantage of reinforcement learning methods that use off-line training with data, as well as on-line training during real-time operation. The new algorithms may also find use in reinforcement learning contexts involving approximation, such as multistep lookahead and tree search schemes, and/or rollout algorithms.

We consider deterministic infinite horizon optimal control problems with nonnegative stage costs. We draw inspiration from learning model predictive control scheme designed for continuous dynamics and iterative tasks, and propose a rollout algorithm that relies on sampled data generated by some base policy. The proposed algorithm is based on value and policy iteration ideas, and applies to deterministic problems with arbitrary state and control spaces, and arbitrary dynamics. It admits extensions to problems with trajectory constraints, and a multiagent structure.

In this paper we aim to provide analysis and insights (often based on visualization), which explain the beneficial effects of on-line decision making on top of off-line training. In particular, through a unifying abstract mathematical framework, we show that the principal AlphaZero/TD-Gammon ideas of approximation in value space and rollout apply very broadly to deterministic and stochastic optimal control problems, involving both discrete and continuous search spaces. Moreover, these ideas can be effectively integrated with other important methodologies such as model predictive control, adaptive control, decentralized control, discrete and Bayesian optimization, neural network-based value and policy approximations, and heuristic algorithms for discrete optimization.