Make sense of data complexity

Abstract the dimensions and granularity of data to understand, clarify assumptions and think logically ecosystem.

Why multiple knobs

Some knobs summarize data, some describe dependencies with other process, Some knobs let you transform data, apply AI and predict, suggest approaches

From the blog

Debugging NN is hard

Testing, generative coverage and debugging is hard

To build a reliable product or mission critical system it is important to test these throughly. Advance AI uses Neural Network, Deep Neural Network (DNN) Q Learning is an algorithm used in hierarchal learning. However Q learning find it hard to learn from very big state space. It does not work in continuous space as exploring all states is hard. Learning policy is even harder. Key to success is - can you create an abstract world (dataset) that let you learn good policies.

Combine VAR and RL agent

Generate data how trajectories will evolve

Generative AI and RL agent can work together to build a robust system. A Variational Auto Encoder (VAE) learn from latent space and then it geenrate future trajectories. An RL agent can learn from these and build an optimal policy. Using knobs one can keep state space limited and focus on abstraction that are required for learning.


Why knobs matter

Discovering abstractions that reduce the amount of experience or thinking time an RL agent requires to find a good solution is key for RL success. Knobs let you control how big or small state space you can use.

Using knobs you can manage a trade off between compression of states and representation of good behavior.