🤖 Implement sim2sim training based on Unitree_rl_gym, enabling motion and policy modification for enhanced robotic behavior in Beyondmimic.

ALX Higher level programming projects are solutions to solve complex challenges

Classification (Pattern Recognition) algorithm development with Bayesian, Anti-Bayesian, Decision Tree and Dependence Tree concepts. [IEMIS 2020, Springer AISC]

Exploration of brain-inspired computing architectures through computational neuroscience models, focusing on point neuron dynamics and spiking neural networks (SNNs) using Nengo. Bridges concepts from neuroscience, computer science, and electrical engineering.

Developed a custom clustering algorithm to analyze wine data without traditional machine learning. The project standardizes features and employs mathematical formulas using NumPy to identify distinct clusters, offering insights into wine sample groupings and their characteristics.

⚡ Optimize memory with RamTorch, a PyTorch library for training large deep learning models efficiently using CPU-GPU hybrid techniques.

⚡ Optimize heterogeneous reinforcement learning with GEPO for decentralized AI training, tackling policy staleness in large models effectively.