DQN Optimization for Atari Breakout

Job ID: 39422008

Budget: £20 – £250 GBP

I'm seeking an expert to help diagnose and fix slow convergence issues in my DQN training on the Atari game Breakout.

Requirements:
- Expertise in Q-learning and DQN
- Proficiency in TensorFlow, Keras, Gymnasium
- Familiarity with CUDA and cuDNN
- Ability to analyze and optimize training parameters

Ideal Skills:
- Strong background in reinforcement learning
- Experience with Atari game environments
- Problem-solving skills to identify training bottlenecks

Your goal will be to determine the root cause of slow convergence and suggest effective solutions to improve training performance. Please provide a detailed plan with your bid.
Related categories: Python Tensorflow Keras