Global ETD Search

Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.

Results

Showing 1 to 20 of 38 for “"Value function approximation"”.

  1. Feature Selection for Value Function Approximation

    … it to reach its goal. This is quantified with value functions, which evaluate a state and return the sum of rewards the agent can expect to receive from that state. Given a good value function, the agent can choose the actions which maximize this sum of rewards. Value functions are often chosen …

    duke Repository record for Feature Selection for Value Function Approximation (opens in a new tab)

  2. Sparse Value Function Approximation for Reinforcement Learning

    … reinforcement learning (RL) algorithms is the approximation of the value function. The design and selection of features for approximation in RL is crucial, and an ongoing area of research. One approach to the problem of feature selection is to apply sparsity-inducing techniques in learning the …

    duke Repository record for Sparse Value Function Approximation for Reinforcement Learning (opens in a new tab)

  3. Value function approximation architectures for neuro-dynamic programming

    … only within a prescribed finite-dimensional function class. Thus, the question that always arises is how should the function class be chosen? In this dissertation, we first propose an approach using the solutions to associated fluid and diffusion approximations. In order to evaluate this …

    uiuc Repository record for Value function approximation architectures for neuro-dynamic programming (opens in a new tab)

  4. Value function approximation methods for Linearly-solvable Markov Decision Process

    … Finding the solution of Bellman equation, the value function, or cost-to-go function, suffers from multiple difficulties, including the curse of dimensionality. In the linearly-solvable Markov Decision Process (LMDP) framework, the Bellman equation can be linearized despite nonlinearity in the …

    washington Repository record for Value function approximation methods for Linearly-solvable Markov Decision Process (opens in a new tab)

  5. Value Function Approximation Algorithms for Reinforcement Learning in Delay-Sensitive Wireless Communications

    … structural properties of the resulting optimal value function, which quantifies the long-run performance of the optimal scheduling policy. We show that the optimal value function (i) is non-decreasing and has increasing differences in the queue backlog; (ii) is non-increasing and has increasing …

    buffalo Repository record for Value Function Approximation Algorithms for Reinforcement Learning in Delay-Sensitive Wireless Communications (opens in a new tab)

  6. Relaxations for the dynamic knapsack problem with stochastic item sizes

    … new semi-infinite relaxation based on an affine value function approximation, and show that an existing pseudo-polynomial relaxation corresponds to a non-parametric value function approximation. We compare both theoretically to other relaxations from the literature and also perform a …

    gatech Repository record for Relaxations for the dynamic knapsack problem with stochastic item sizes (opens in a new tab)

  7. Approximate dynamic programming methods for advance patient scheduling

    … a linear programming method based on an affine value function approximation. This thesis starts by presenting an alternative solution approach for this problem based on the use of simulation, a policy iteration framework and a non-linear value function approximation. It then extends the dynamic …

    ubc Repository record for Approximate dynamic programming methods for advance patient scheduling (opens in a new tab) Download PDF for Approximate dynamic programming methods for advance patient scheduling (opens in a new tab)

  8. Application of temporal difference learning to the game of Snake.

    … computer controlled players is documented. Using value function approximation with multilayer artificial neural networks and the actor-critic architecture, computer players capable of playing the game of Snake can be created. The adaptation to the standard neural network backpropagation procedure …

    louisville Repository record for Application of temporal difference learning to the game of Snake. (opens in a new tab)

  9. Advances in Tactical & Operational Planning for Less-than-Truckload Carriers

    … in Chapter 4, we describe a lookup table value function approximation (LT-VFA) mechanism for this ADP algorithm, and introduce and compare a number of aggregation approaches which use features of the post-decision states to aggregate the post-decision state space. Furthermore, since the …

    gatech Repository record for Advances in Tactical & Operational Planning for Less-than-Truckload Carriers (opens in a new tab)

  10. Efficient Data-Driven Robust Policies for Reinforcement Learning

    … and robust feature selection method for linear value function approximation, a standard approach to solving reinforcement learning problems with large state spaces. Our experiments show that our technique is faster and more stable than alternative methods.

    unh-thes Repository record for Efficient Data-Driven Robust Policies for Reinforcement Learning (opens in a new tab)

  11. Neural Network Approximations for Mathematical Optimization

    … in each under a general framework based on value function approximation and optimizing over trained neural networks. Then, in the subsequent chapters, we present tailored deep learning architectures and surrogate optimization formulations that enable efficient decision-making in stochastic …

    toronto-retro Repository record for Neural Network Approximations for Mathematical Optimization (opens in a new tab)

  12. New Optimization Models and Algorithms for Logistics and Delivery Problems

    … to make delivery. We develop a policy based on value function approximation (VFA) to determine delivery fees and provide feasible vehicle routes jointly in real-time. After that, we show how VFA-based future anticipation improves the total profit and increases the number of served customers.

    umn Repository record for New Optimization Models and Algorithms for Logistics and Delivery Problems (opens in a new tab)

  13. Stochastic decision models for last mile distribution using approximate dynamic programming

    … to our prior work. First, we evaluate two Value Function Approximation methods and find that Rollout Algorithms can serve as a proxy for the optimal solution. Against this as a benchmark, CD performs well when the amount of items donated greatly exceeds the demand, and also when demand …

    alabama Repository record for Stochastic decision models for last mile distribution using approximate dynamic programming (opens in a new tab)

  14. Generation Capacity Expansion in Restructured Energy Markets

    … planning (GEP) model, we first develop a novel value function approximation based reinforcement learning (RL) algorithm. Currently there exist only mathematical programming based solution approaches for two player games and the N-player extensions in literature still have several unresolved …

    usf Repository record for Generation Capacity Expansion in Restructured Energy Markets (opens in a new tab)

  15. Sparse Bayesian reinforcement learning

    … part of reinforcement learning, and it requires function approximation. However, with function approximation, the most popular TD methods such as TD(λ), SARSA, and Q-learning lose stability and diverge especially when the complexity of the problem grows and the sampling distribution is biased. …

    colostate Repository record for Sparse Bayesian reinforcement learning (opens in a new tab)

  16. Electric Vehicle Fleet Charging Management

    … to replace the charging decisions' expected value. It employs a value function approximation, enabling rapid charging solutions, even for large fleets, with competitive costs and service levels. Lastly, the dissertation examines a decision support system (DSS) integrating optimal stopping …

    alabama Repository record for Electric Vehicle Fleet Charging Management (opens in a new tab)

  17. Strategic, Tactical, and Operational Decision-making in Crowd-shipping

    … The numerical experiments demonstrate the value of the parcel locker location solution from the proposed model compared to the locations obtained from the customer pickup model. The second problem proposes a novel two-tier delivery system with crowd-shipping and mobile depots, considering …

    toronto-retro Repository record for Strategic, Tactical, and Operational Decision-making in Crowd-shipping (opens in a new tab)

  18. Adaptive Optimal Control via Reinforcement Learning: Theory and Its Application to Automotive Engine Systems

    … knowledge. The finite-time convergence of the value function approximation is guaranteed by using a sliding-mode technique while the persistent excitation (PE) condition of the state trajectories can be verified directly in real time. Furthermore, the proposed Q-learning approach is extended to …

    bristol Repository record for Adaptive Optimal Control via Reinforcement Learning: Theory and Its Application to Automotive Engine Systems (opens in a new tab) Download PDF for Adaptive Optimal Control via Reinforcement Learning: Theory and Its Application to Automotive Engine Systems (opens in a new tab)

  19. Quantitative temporal logic for safe and robust planning, learning, and control

    … operates over continuous-time signals with real-valued predicates and has established quantitative robustness measures, while Time Window Temporal Logic (TWTL) expresses sequential tasks with bounded time horizons through explicit time windows but previously lacked quantitative semantics. …

    bu Repository record for Quantitative temporal logic for safe and robust planning, learning, and control (opens in a new tab)

Page 1 of 2