Sökning: "Actor-Critic"
Visar resultat 1 - 5 av 22 uppsatser innehållade ordet Actor-Critic.
1. Scalable Reinforcement Learning for Linear-Quadratic Control of Networks
Uppsats för yrkesexamina på avancerad nivå, Lunds universitet/Institutionen för reglerteknikSammanfattning : Distributed optimal control is known to be challenging and can become intractable even for linear-quadratic regulator problems. In this work, we study a special class of such problems where distributed state feedback controllers can give near-optimal performance. LÄS MER
2. Scalable Reinforcement Learning for Formation Control with Collision Avoidance : Localized policy gradient algorithm with continuous state and action space
Master-uppsats, KTH/Skolan för teknikvetenskap (SCI); KTH/Skolan för elektroteknik och datavetenskap (EECS)Sammanfattning : In the last decades, significant theoretical advances have been made on the field of distributed mulit-agent control theory. One of the most common systems that can be modelled as multi-agent systems are the so called formation control problems, in which a network of mobile agents is controlled to move towards a desired final formation. LÄS MER
3. Intelligent autoscaling in Kubernetes : the impact of container performance indicators in model-free DRL methods
Master-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)Sammanfattning : A key challenge in the field of cloud computing is to automatically scale software containers in a way that accurately matches the demand for the services they run. To manage such components, container orchestrator tools such as Kubernetes are employed, and in the past few years, researchers have attempted to optimise its autoscaling mechanism with different approaches. LÄS MER
4. Deep Reinforcement Learning Approach to Portfolio Optimization
Kandidat-uppsats, Lunds universitet/Nationalekonomiska institutionenSammanfattning : This paper evaluates whether a deep reinforcement learning (DRL) approach can be implemented, on the Swedish stock market, to optimize a portfolio. The objective is to create and train two DRL algorithms that can construct portfolios that will be benchmarked against the market portfolio, tracking OMXS30, and the two conventional methods, the naive portfolio, and minimum variance portfolio. LÄS MER
5. Spiking Reinforcement Learning for Robust Robot Control Under Varying Operating Conditions
Master-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)Sammanfattning : Over the last few years, deep reinforcement learning (RL) has gained increasing popularity for its successful application to a variety of complex control and decision-making tasks. As the demand for deep RL algorithms deployed in challenging real-world environments grows, their robustness towards uncertainty, disturbances and perturbations of the environment becomes more and more important. LÄS MER