Sökning: "Förstärkande inlärning"

Visar resultat 16 - 20 av 35 uppsatser innehållade orden Förstärkande inlärning.

  1. 16. A Graph Attention plus Reinforcement Learning Method for Antenna Tilt Optimization

    Master-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)

    Författare :Tengfei Ma; [2021]
    Nyckelord :Graph Attention; Reinforcement Learning; Antenna Tilt Optimization; 5G; Attention Mechanism; Graph; DQN; Back- Propagation; Gradient Descent;

    Sammanfattning : Remote Electrical Tilt optimization is an effective method to obtain the optimal Key Performance Indicators (KPIs) by remotely controlling the base station antenna’s vertical tilt. To improve the KPIs aims to improve antennas’ cooperation effect since KPIs measure the quality of cooperation between the antenna to be optimized and its neighbor antennas. LÄS MER

  2. 17. Model Based Reinforcement Learning for Automatic Tuning of Cavity Filters

    Master-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)

    Författare :Doumitrou Daniil Nimara; [2021]
    Nyckelord :Reinforcement Learning; Cavity Filter Tuning; Sample Complexity; Background Planning; Förstärkande inlärning; inställning av kavitetsfilter; provkomplexitet; bakgrundsplanering;

    Sammanfattning : As telecommunication continues developing, the demand for mass production of well calibrated Base Transceiver Stations (BTS) components increases. Cavity Filters are an essential piece of every BTS; however, manufacturing tolerances often lead to detuned filters which require costly post-production fine tuning. LÄS MER

  3. 18. Belief-aided Robust Control for Remote Electrical Tilt Optimization

    Master-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)

    Författare :Jack Jönsson; [2021]
    Nyckelord :Mobile Network Optimization; Remote Electrical Tilt; Robust Control; Reinforcement Learning; Partially Observable Markov Decision Process; Belief State Estimation; Bayesian Neural Network; Optimering av Mobilnätverk; Fjärrstyrning av Elektrisk Lutning; Robust Reglering; Delvis Observerbar Markovprocess; Tillståndsestimering; Bayesiskt Neuralt Nätverk;

    Sammanfattning : Remote Electrical Tilt (RET) is a method for configuring antenna downtilt in base stations to optimize mobile network performance. Reinforcement Learning (RL) is an approach to automating the process by letting an agent learn an optimal control strategy and adapt to the dynamic environment. LÄS MER

  4. 19. Warehouse Optimization by Multi-Agent Rollout Algorithms

    Kandidat-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)

    Författare :Laura Briffa; William Emanuelsson; [2021]
    Nyckelord :multi-agent problems; reinforcement learning; optimization and optimal control; collision avoidance; warehouse;

    Sammanfattning : Systems consisting of multiple robots are traditionallydifficult to optimize. This project considers such a systemin a simulated warehouse setting, where the robots are todeliver boxes while avoiding collisions. Adding such collisionconstraints complicates the problem. LÄS MER

  5. 20. Parameter Tuning in a Jet Printing Machine usingReinforcement Learning

    Master-uppsats, KTH/Skolan för industriell teknik och management (ITM)

    Författare :ALEXANDER MURTAZA; [2021]
    Nyckelord :Jet printer; Reinforcement Learning; Q-learning; Parameter tuning; Surface mount technology; Jet printer; Förstärkande inlärning; Q-learning; Parameter justering; Ytmontering;

    Sammanfattning : Surface mount technology is a common way to assembly electrical components onto PrintedCircuit Boards (PCB). To assemble the components, solder paste is used. One way to apply solderpaste onto PCB is jet printing. LÄS MER