Sökning: "Axel Jernbäcker"

Hittade 3 uppsatser innehållade orden Axel Jernbäcker.

1. 1. Kalman filters as an enhancement to object tracking using YOLOv7

   Master-uppsats, KTH/Matematik (Avd.)

   Författare :Axel Jernbäcker; [2022]
   Nyckelord :statistics; applied mathematics; machine learning; kalman filter; statistik; tillämpad matematik; maskinlärning; kalman filter;

   Sammanfattning : In this paper we study continuous tracking of airplanes using object detection models, namely YOLOv7, combined with a Kalman filter. The tracking should be able to be done in real-time. The idea of combining Kalman filters with an object detection model comes from the lack of time-dependent context in models such as YOLOv7. LÄS MER

3. 2. Training AI for asset pricing

   C-uppsats, Handelshögskolan i Stockholm/Institutionen för finansiell ekonomi

   Författare :Axel Jernbäcker; Henrik Angelstig; [2022]
   Nyckelord :AI; Machine learning; Asset pricing; stock price prediction; time-weighting;

   Sammanfattning : This bachelor's thesis examines how a machine learning program's ability to predict stock prices is affected by how the data on which the program is trained has been time-weighted. Building upon the work by Gu, Kelly, and Xiu (2020), we examine how the same machine learning program performs when its training data is time-weighted using one of three methods: exponential weighting, linear weighting, or cutoff. LÄS MER

4. 3. Search methods for Strategy Synthesis for Multi-Agent Games of Imperfect Information

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

   Författare :Axel Jernbäcker; Max Junestrand; [2021]
   Nyckelord :;

   Sammanfattning : Finding strategies for agents in multi-agent systems of imperfect information is a difficult and time-consuming endeavor. It is thus of interest to find ways of doing this more effectively. This study aims to improve the performance of a strategy synthesizer tool by testing two search techniques and evaluating their performance. LÄS MER