arXiv subjects

Yuchao Li

Publications and source records attributed to Yuchao Li.

At least 19 recordsLinked to original sources

arXiv2026

Recovery Control in Replicated Systems through Autonomous Multiagent Rollout

We study recovery control in replicated computing systems. Such systems consist of replicas that collectively provide a service to a client population. This redundancy enables the system to withstand failures provided that failed replicas are recovered faster than new failures occur. We show that the problem of deciding when to initiate recovery of selected replicas can be formulated as a partially observable Markov decision problem (POMDP) with a multiagent structure. We exploit this structure to apply a multiagent rollout method for approximating optimal control policies. Our method uses precomputed signaling information that reduces the need for replica coordination and facilitates parallel computations. Experiments show that our method scales to systems with up to 70 replicas and reduces costs compared to the recovery policies currently used in practice.

eess.SY↗

arXiv2026

Compressed Traffic Assignment with the Augmented Lagrangian Method

We consider large-scale traffic assignment problems and develop a path-based compression framework. In particular, we partition paths into major and minor paths according to a set of nominal flows and a prescribed threshold, and retain the major paths explicitly. For the minor paths, we introduce a low-dimensional representation based on a truncated singular value decomposition of the minor path-link incidence matrix. We also provide a feasibility safeguard that ensures the compressed problem remains feasible. To solve the resulting formulation, we use an augmented Lagrangian method with separate penalty parameters for the different constraints and adaptive penalty parameter updates. We report computational studies using the Chicago Sketch, Chicago Regional, and Philadelphia networks. The results show a compression-accuracy trade-off: moderate thresholds can achieve substantial dimension reduction while maintaining high link-flow fidelity, whereas more aggressive compression tends to increase iteration counts and objective gaps. In our rank-sensitivity experiments, increasing the compression rank beyond moderate values produces little improvement in solution quality while increasing computational cost substantially. Overall, the proposed framework offers a practical way to reduce the dimensionality of large path-based traffic assignment problems while preserving feasibility and good solution quality.

math.OC↗

arXiv2026

On-Line Policy Iteration with Trajectory-Driven Policy Generation

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 {and a robot arm} 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.

eess.SY↗

arXiv2026

Model Predictive Control For Trade Execution

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% relative to spread-crossing benchmarks and achieves significant reductions in slippage. Moreover, augmenting the base policy with predictive price information further enhances performance, highlighting the framework's flexibility for integration with forecasting components.

q-fin.TR↗

arXiv2025

Adaptive Network Security Policies via Belief Aggregation and Rollout

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 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. The method uses a model or simulator of the system, which is updated when changes occur, and combines three components: belief estimation through particle filtering, offline policy computation through feature-based aggregation, and online policy adaptation through rollout. In particular, feature-based aggregation enables scalable offline optimization of a policy, while rollout adapts the policy online to changes in the system model without repeating the offline optimization. We analyze the approximation error of the aggregation and show that the 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.

eess.SY↗

arXiv2025

Feature-Based Belief Aggregation for Partially Observable Markov Decision Problems

We consider a finite-state partially observable Markov decision problem (POMDP) with an infinite horizon and a discounted cost, and we propose a new method for computing a cost function approximation that is based on features and aggregation. In particular, using the classical belief-space formulation, we construct a related Markov decision problem (MDP) by first aggregating the unobservable states into feature states, and then introducing representative beliefs over these feature states. This two-stage aggregation approach facilitates the use of dynamic programming methods for solving the aggregate problem and provides additional design flexibility. The optimal cost function of the aggregate problem can in turn be used within an on-line approximation in value space scheme for the original POMDP. We derive a new bound on the approximation error of our scheme. In addition, we establish conditions under which the cost function approximation provides a lower bound for the optimal cost. Finally, we present a biased aggregation approach, which leverages an optimal cost function estimate to improve the quality of the approximation error of the aggregate problem.

eess.SY↗

arXiv2025

An Error Bound for Aggregation in Approximate Dynamic Programming

We consider a general aggregation framework for discounted finite-state infinite horizon dynamic programming (DP) problems. It defines an aggregate problem whose optimal cost function can be obtained off-line by exact DP and then used as a terminal cost approximation for an on-line reinforcement learning (RL) scheme. We derive a bound on the error between the optimal cost functions of the aggregate problem and the original problem. This bound was first derived by Tsitsiklis and van Roy [TvR96] for the special case of hard aggregation. Our bound is similar but applies far more broadly, including to soft aggregation and feature-based aggregation schemes.

math.OC↗

arXiv2025

Distributed Charging Coordination for Electric Trucks under Limited Facilities and Travel Uncertainties

In this work, we address the problem of charging coordination between electric trucks and charging stations. The problem arises from the tension between the trucks' nontrivial charging times and the stations' limited charging facilities. Our goal is to reduce the trucks' waiting times at the stations while minimizing individual trucks' operational costs. We propose a distributed coordination framework that relies on computation and communication between the stations and the trucks, and handles uncertainties in travel times and energy consumption. Within the framework, the stations assign a limited number of charging ports to trucks according to the first-come, first-served rule. In addition, each station constructs a waiting time forecast model based on its historical data and provides its estimated waiting times to trucks upon request. When approaching a station, a truck sends its arrival time and estimated arrival-time windows to the nearby station and the distant stations, respectively. The truck then receives the estimated waiting times from these stations in response, and updates its charging plan accordingly while accounting for travel uncertainties. We performed simulation studies for $1,000$ trucks traversing the Swedish road network for $40$ days, using realistic traffic data with travel uncertainties. The results show that our method reduces the average waiting time of the trucks by $46.1\%$ compared to offline charging plans computed by the trucks without coordination and update, and by $33.8\%$ compared to the coordination scheme assuming zero waiting times at distant stations.

eess.SY↗

arXiv2025

Semilinear Dynamic Programming: Analysis, Algorithms, and Certainty Equivalence Properties

We consider a broad class of dynamic programming (DP) problems that involve a partially linear structure and some positivity properties in their system equation and cost function. We address deterministic and stochastic problems, possibly with Markov jump parameters. We focus primarily on infinite horizon problems and prove that under our assumptions, the optimal cost function is linear, and that an optimal policy can be computed efficiently with standard DP algorithms. Moreover, we show that forms of certainty equivalence hold for our stochastic problems, in analogy with the classical linear quadratic optimal control problems.

math.OC↗

arXiv2024

Superior Computer Chess with Model Predictive Control, Reinforcement Learning, and Rollout

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.

cs.AI↗

arXiv2024

An Approximate Dynamic Programming Framework for Occlusion-Robust Multi-Object Tracking

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% improvement in the 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.

cs.CV↗

arXiv2024

Most Likely Sequence Generation for $n$-Grams, Transformers, HMMs, and Markov Chains, by Using Rollout Algorithms

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.

cs.LG↗

arXiv2024

Distributed Charging Coordination of Electric Trucks with Limited Charging Resources

Electric trucks usually need to charge their batteries during long-range delivery missions, and the charging times are often nontrivial. As charging resources are limited, waiting times for some trucks can be prolonged at certain stations. To facilitate the efficient operation of electric trucks, we propose a distributed charging coordination framework. Within the scheme, the charging stations provide waiting estimates to incoming trucks upon request and assign charging ports according to the first-come, first-served rule. Based on the updated information, the individual trucks compute where and how long to charge whenever approaching a charging station in order to complete their delivery missions timely and cost-effectively. We perform empirical studies for trucks traveling over the Swedish road network and compare our scheme with the one where charging plans are computed offline, assuming unlimited charging facilities. It is shown that the proposed scheme outperforms the offline approach at the expense of little communication overhead.

eess.SY↗

arXiv2024

Rollout-Based Charging Strategy for Electric Trucks with Hours-of-Service Regulations (Extended Version)

Freight drivers of electric trucks need to design charging strategies for where and how long to recharge the truck in order to complete delivery missions on time. Moreover, the charging strategies should be aligned with drivers' driving and rest time regulations, known as hours-of-service (HoS) regulations. This letter studies the optimal charging problems of electric trucks with delivery deadlines under HoS constraints. We assume that a collection of charging and rest stations is given along a pre-planned route with known detours and that the problem data are deterministic. The goal is to minimize the total cost associated with the charging and rest decisions during the entire trip. This problem is formulated as a mixed integer program with bilinear constraints, resulting in a high computational load when applying exact solution approaches. To obtain real-time solutions, we develop a rollout-based approximate scheme, which scales linearly with the number of stations while offering solid performance guarantees. We perform simulation studies over the Swedish road network based on realistic truck data. The results show that our rollout-based approach provides near-optimal solutions to the problem in various conditions while cutting the computational time drastically.

eess.SY↗

arXiv2023

Exact Dynamic Programming for Positive Systems with Linear Optimal Cost

Recent work [Ran22] formulated a class of optimal control problems involving positive linear systems, linear stage costs, and elementwise constraints on control. It was shown that the problem admits linear optimal cost and the associated Bellman's equation can be characterized by a finite-dimensional nonlinear equation, which is solved by linear programming. In this work, we report exact dynamic programming (DP) theories for the same class of problems. Moreover, we extend the results to a related class of problems where the norms of control are bounded while the optimal costs remain linear. In both cases, we provide conditions under which the solutions are unique, investigate properties of the optimal policies, study the convergence of value iteration, policy iteration, and optimistic policy iteration applied to such problems, and analyze the boundedness of the solution to the associated optimization programs. Apart from a form of the Frobenius-Perron theorem, the majority of our results are built upon generic DP theory applicable to problems involving nonnegative stage costs.

math.OC↗

arXiv2023

Parallel Model Predictive Control for Deterministic Systems

In this note, we consider infinite horizon optimal control problems with deterministic systems. Since exact solutions to these problems are often intractable, we propose a parallel model predictive control (MPC) method that provides an approximate solution. Our method computes multiple lookahead minimization problems at each time, where each minimization may involve a different number of lookahead steps, and terminal cost and constraint. The policy computed via parallel MPC applies the first control of the lookahead minimization with the lowest cost. We show that the proposed method can harnesses the power of multiple computing units. Moreover, we prove that the policy computed via parallel MPC has better performance guarantee than that computed via the single lookahead minimization involved in parallel MPC.

math.OC↗

arXiv2023

Flash-LLM: Enabling Cost-Effective and Highly-Efficient Large Generative Model Inference with Unstructured Sparsity

With the fast growth of parameter size, it becomes increasingly challenging to deploy large generative models as they typically require large GPU memory consumption and massive computation. Unstructured model pruning has been a common approach to reduce both GPU memory footprint and the overall computation while retaining good model accuracy. However, the existing solutions do not provide a highly-efficient support for handling unstructured sparsity on modern GPUs, especially on the highly-structured Tensor Core hardware. Therefore, we propose Flash-LLM for enabling low-cost and highly-efficient large generative model inference with the sophisticated support of unstructured sparsity on high-performance but highly restrictive Tensor Cores. Based on our key observation that the main bottleneck of generative model inference is the several skinny matrix multiplications for which Tensor Cores would be significantly under-utilized due to low computational intensity, we propose a general Load-as-Sparse and Compute-as-Dense methodology for unstructured sparse matrix multiplication. The basic insight is to address the significant memory bandwidth bottleneck while tolerating redundant computations that are not critical for end-to-end performance on Tensor Cores. Based on this, we design an effective software framework for Tensor Core based unstructured SpMM, leveraging on-chip resources for efficient sparse data extraction and computation/memory-access overlapping. At SpMM kernel level, Flash-LLM significantly outperforms the state-of-the-art library, i.e., Sputnik and SparTA by an average of 2.9x and 1.5x, respectively. At end-to-end framework level on OPT-30B/66B/175B models, for tokens per GPU-second, Flash-LLM achieves up to 3.8x and 3.6x improvement over DeepSpeed and FasterTransformer, respectively, with significantly lower inference cost.

cs.DC↗

arXiv2023

Performance Bounds of Model Predictive Control for Unconstrained and Constrained Linear Quadratic Problems and Beyond

We study unconstrained and constrained linear quadratic problems and investigate the suboptimality of the model predictive control (MPC) method applied to such problems. Considering MPC as an approximate scheme for solving the related fixed point equations, we derive performance bounds for the closed-loop system under MPC. Our analysis, as well as numerical examples, suggests new ways of choosing the terminal cost and terminal constraints, which are \emph{not} related to the solution of the Riccati equation of the original problem. The resulting method can have a larger feasible region, and cause hardly any loss of performance in terms of the closed-loop cost over an infinite horizon.

math.OC↗