AlphaZero Implementation with Monte Carlo Tree Search for Tic-Tac-Toe and Connect Four Developed a simplified AlphaZero-style reinforcement learning framework integrating neural networks with Monte Carlo Tree Search (MCTS). Implemented self-play training to learn optimal strategies from scratch and applied the system to Tic-Tac-Toe and Connect Four, demonstrating policy and value network training, game state evaluation, and search-based decision making.