Dr. Dirichukwu Goodluck Oguzie
Researcher & Lecturer
PhD, MSc, BSc • Computer Science
Specializing in RL & Social Robotics
I develop intelligent systems that integrate reinforcement learning, predictive models, and neural network architectures to solve real-world problems in social robotics. My research focuses on creating safer, more efficient autonomous navigation systems, with work published at top-tier conferences including IEEE PRML 2024 and UKCI 2023.
Featured Publications
Cosine-Gated LSTM for Sequential Learning
Novel Cosine-Gated LSTM architecture for improved sequential learning, outperforming LSTM, GRU, and Transformer models.
Predictive World Models for Social Navigation
Predictive models (2StepAhead, MASPM) for safe autonomous navigation, improving performance and reducing collisions.
Current Research
Adaptive Prediction Horizons in Reinforcement Learning
Performance vs efficiency: adaptive prediction horizon achieves highest return among low-latency models
Entropy-driven adaptive prediction horizons for RL, enabling dynamic adjustment based on environmental complexity.
Wi-Fi CSI Action Recognition with ESP32
Privacy-preserving action recognition using wireless signals
Wi-Fi CSI action recognition with ESP32 for privacy-preserving human action prediction using wireless signals.
Teaching
I teach AI and machine learning courses with 13+ years of experience at multiple institutions, integrating cutting-edge research with practical application.
Open for Collaboration
Seeking research partnerships in Reinforcement Learning & Autonomous Systems, Social Robot Navigation, and Neural Network Architectures.
Open to joint projects, student co-supervision, industry partnerships, and grant applications.