Reinforcement learning and windlass: what they have in common

1. Step 1. Reinforcement learning — uses — Artificial neural network. Deep RL.
2. Step 2. Artificial neural network — uses — Algorithm. Training is the backpropagation algorithm.
3. Step 3. Algorithm — is used in — Rubik’s Cube. Solved with algorithms.
4. Step 4. Rubik’s Cube — has part — Spring. Springs adjust the tension.
5. Step 5. Spring — is part of — Ratchet. The pawl spring.
6. Step 6. Ratchet — is part of — Windlass. A ratchet holds the load.

Open Evothings