TeardexRobots/Parts/Training
Reinforcement learning
Learning by trial and error. The robot, usually a virtual copy of it, tries something, gets a score for how well it went, and adjusts. After millions of tries it has a skill nobody wrote down, such as walking over rough ground.
- 1 of 5
- robots in the database
- Not named
- made by
- Training
- brain layer
- Not public
- price
Key numbers
- Used on NEO for
- walking and moving about
Source: 1X, NEO AI page (1x.tech)
Strengths and trade-offs
Our analysisStrengths
Finds ways of moving that people would not think to program.
Produces sturdy walking that recovers from pushes and slips.
Can be repeated cheaply in simulation.
Trade-offs
Needs a simulator close enough to reality.
Hard to know why the robot does what it does.
Good at movement, much weaker at tasks that need understanding.
Why robot makers choose it
Our readBalance and walking are too quick and too complicated to write by hand. Practising millions of times in simulation gives a robot steadier legs than any written rules have managed.
This is our reasoning from the public facts. No maker has confirmed it.
Robots using it
This list covers only the robots researched so far. A robot missing here may still use this part.
Numbers and listed prices on this page come from the sources shown. Anything marked as our analysis, our read or our estimate is our own view and has not been confirmed by any company. Last checked 12 Oct 2026. How we verify