Visualizing hyperparameters with a robot arm
Channel:
Subscribers:
9,890
Published on ● Video Link: https://www.youtube.com/watch?v=_Me2qhX9jNo
Using Gradient Descent to solve the Inverse Kinematics for a robot arm lets us visualize the impact that hyperparameters have on the optimization.
Too high a learning rate and we start thrashing. Too low and we converge too fast. Too much momentum and we circle the target if we approach too quickly. Just the right amount of momentum more than doubles how fast we converge at the optimal solution.
Other Videos By Seabass
2023-03-06 | AI-Arena Quickstart - Creating your first AI |
2023-02-27 | AI-Arena Devlog 1 |
2023-02-16 | Procedural Galaxy in Three.js |
2023-02-15 | Let's Code -- Procedural Galaxy in Javascript |
2023-02-13 | Let's Code - Procedural Galaxy in Javascript |
2023-02-02 | Inverse Kinematics with Keyframing Timeline |
2023-01-25 | Inverse Kinematics with Obstacle Avoidance in Webassembly |
2023-01-18 | Inverse Kinematics with Obstacle Avoidance test |
2023-01-07 | IKSolverHybrid |
2023-01-03 | 3D Inverse Kinematics in Three.js |
2022-12-20 | Visualizing hyperparameters with a robot arm |
2022-11-16 | 3D Foxhole Map |
2022-09-04 | I made a game where you code your own AI |
2022-07-18 | GMTK 2022 submission |
2022-05-01 | Generating Music with AI |
2022-01-16 | Procedural Walking |
2021-10-11 | Shooter 4 |
2021-08-27 | Unity Procedural Walk |
2021-08-11 | Unity Flowfield Utility |
2021-04-13 | 0-100+ hours of practice Johnny B. Goode intro |
2021-01-13 | Unity RTS after 2 weeks |