ml-rubiks

A machine-learned, human-understandable method for solving the Rubik's Cube using strictly algorithmic logic.

Method Goals

1. <= 4 Steps: Clearly defined sequence.
2. < 50 Algorithms: Minimize memorization while maintaining speed.
3. Zero Intuition: Replaces "intuitive" F2L with deterministic search-and-execute triggers.
4. Sub-20s Average: Optimized for high-speed execution under specific constraints.

The Time Efficiency Model

To satisfy the 20-second goal, the method must perform within the following parameters:

T_{total} = \frac{M}{y} + ((S - 1) \cdot x) + (z \cdot p)

Where: (in parenthesis ideal values)

• M: Average move count (~75)
• y: Turns per second (6.2 TPS)
• S: Number of steps (4)
• x: Transition delay (0.35s)
• z: Mistakes (1.5)
• p: Penalty per correction (1.0s)