A5043053491- Energy Load and Power Forecasting
- Stock Market Forecasting Methods
- Complex Systems and Time Series Analysis
- Smart Grid Energy Management
- Advanced Measurement and Detection Methods
- Computational Physics and Python Applications
- Time Series Analysis and Forecasting
- Anomaly Detection Techniques and Applications
- Industrial Technology and Control Systems
- Auction Theory and Applications
- Advanced Sensor and Control Systems
- Electric Power System Optimization
- Financial Markets and Investment Strategies
Universidade Federal do Paraná
2021-2024
Deep reinforcement learning (DRL) achieved significant progress in several areas enabling computers to perform complex decision-making tasks. Applied quantitative trading, DRL trading agents can optimize their decisions during distinct market scenarios reach a profitable strategy by from previous experiences. This work proposes system containing variant of the Q-Network (DQN) algorithm as its main component called Extended Trading DQN (ETDQN). It adapt trade across numerous market-behavior...
Anomalies are patterns in data that do not conform to a well-defined notion of normal behavior. Anomaly detection has been applied many problems such as bank fraud, fault detection, noise reduction, among others. Some approaches detect anomalies include classical statistical econometric methods AutoRegressive Moving Average (ARMA) and Integrated (ARIMA) approaches. More recently, with the progress artificial intelligence more specifically, machine learning, new algorithms one-class support...