Barry Nichols

Dr. Barry Nichols is a Lecturer in the Department of Computer Science at the University of Hertfordshire, affiliated with the School of Physics, Engineering & Computer Science and the Biocomputation Research Group. With qualifications BSc, PhD, and professional memberships FHEA and MBCS, his research focuses on computational intelligence techniques.

• University: University of Hertfordshire
• School: School of Physics, Engineering & Computer Science
• Department: Department of Computer Science
• Academic Rank: Lecturer
• Professional Memberships: Fellow of the Higher Education Academy (FHEA), Member of the British Computer Society (MBCS)

His work investigates reinforcement learning, artificial neural networks, and evolutionary algorithms for solving continuous action-space control problems. Research outputs include applications to the Acrobot system, action selection methods, and Newton's method in dynamic environments.

Recent publications analyze:

• Genetic programming for minimum-time swing-up control
• Eligibility traces in continuous state-action spaces
• Comparison of action selection strategies
• Application of Newton's method to policy optimization

These studies demonstrate expertise in reinforcement learning algorithms, particularly for robotic control systems requiring complex decision-making in continuous domains.

Scientific contributions include:

• Developing computational intelligence techniques for control problems
• Advancing reinforcement learning applications in action-space optimization
• Exploring hybrid approaches combining genetic programming with gradient methods