Name: Alvaro Cabrejas Egea
Type: User
Company: University of Warwick, Fujitsu Research of Europe
Bio: Data Scientist.
Reinforcement Learning and Control, Time Series, Incident Detection.
Currently with @alan-turing-institute, previously with @vivacitylabs.
Location: London, UK
Blog: https://warwick.ac.uk/fac/sci/mathsys/people/students/2015intake/cabrejas-egea/
Alvaro Cabrejas Egea's Projects
A Multi-Agent Reinforcement Learning Environment for Large Scale City Traffic Scenario
Achieving Global Cooperation in Social Networks Through Peer Punishment
Repo for the Deep Reinforcement Learning Nanodegree program
Deviations from Profiles in UK Motorways - Thales, Highways UK and University of Warwick
A toolkit for developing and comparing reinforcement learning algorithms.
Published in IEEE ITSC2018. Paper on generation of Traffic Profiles with Spectral Analysis and Non-Parametric Regression.
Presented in IEEE ITSC2020 (Sep'20) Pre-print version and code. WARP: Wavelet Augmented Regression Profiling
Presented in IEEE 23rd International Conference in Systems, Man, and Cybernetics. Pre-print version of "Assessment of Reward Functions for Reinforcement Learning Vehicular Traffic Signal Control Under Real-World Limitations"
Pre-print version of "Assessment of Reward Functions in Reinforcement Learning for Multi-Modal Urban Traffic Control under Real-World limitations"
Machine Learning From Scratch. Bare bones NumPy implementations of machine learning models and algorithms with a focus on accessibility. Aims to cover everything from linear regression to deep learning.
Optimisation of computationally intensive noisy simulations with Optimal Stochastic Annealing
Travel Time Profile Estimations using seasonality extraction via non-parametric regression models applied to UK NTIS Motorway data
Comparison between Reward Functions for Urban Traffic Control RL in a real-world intersection.
Jupyter Notebooks for the Python Data Science Handbook
Looking into the relevance of \gamma (discount factor) in RL applied to Urban Traffic Control, looking at the relationship between it and how the timescales in which the state variables vary
VISSIM Reinforcement Learning for Urban Traffic Control. DQN, DDQN, D3QN, A2C, PER.