Deep Active Inference for Pixel-Based Discrete Control: Evaluation on the Car Racing Problem
This work addresses visual-based control for autonomous systems, but it is incremental as it shows limitations compared to existing methods.
The paper tackled the challenge of applying deep active inference to pixel-based discrete control without access to the car's state, achieving performance comparable to deep Q-learning but not reaching state-of-the-art levels.
Despite the potential of active inference for visual-based control, learning the model and the preferences (priors) while interacting with the environment is challenging. Here, we study the performance of a deep active inference (dAIF) agent on OpenAI's car racing benchmark, where there is no access to the car's state. The agent learns to encode the world's state from high-dimensional input through unsupervised representation learning. State inference and control are learned end-to-end by optimizing the expected free energy. Results show that our model achieves comparable performance to deep Q-learning. However, vanilla dAIF does not reach state-of-the-art performance compared to other world model approaches. Hence, we discuss the current model implementation's limitations and potential architectures to overcome them.