https://www.robinleman.com
Code: github.com/RobinLmn/tin...
Code: github.com/RobinLmn/tin...
Code: github.com/RobinLmn/car...
Code: github.com/RobinLmn/car...
Code: github.com/RobinLmn/car...
Code: github.com/RobinLmn/car...
In a CartPole environment using REINFORCE, SGD with momentum averages 490 reward in just 5,000 steps, compared to 292 after 10,000 steps without momentum.
Code: github.com/RobinLmn/car...
In a CartPole environment using REINFORCE, SGD with momentum averages 490 reward in just 5,000 steps, compared to 292 after 10,000 steps without momentum.
Code: github.com/RobinLmn/car...
Couldn't have made it without it.
Couldn't have made it without it.
Using a running baseline not only improves overall agent performance, but also reduces reward variance throughout training.
Code: github.com/RobinLmn/car...
Using a running baseline not only improves overall agent performance, but also reduces reward variance throughout training.
Code: github.com/RobinLmn/car...
🌌 Code: github.com/RobinLmn/deb...
#graphics #rendering #cpp #physics
🌌 Code: github.com/RobinLmn/deb...
#graphics #rendering #cpp #physics
Check it out: github.com/RobinLmn/cud...
#cpp #raytracing #graphics #rendering #hdr
Check it out: github.com/RobinLmn/cud...
#cpp #raytracing #graphics #rendering #hdr
Code: github.com/RobinLmn/cud...
#graphics #cpp #rendering #raytracing
Code: github.com/RobinLmn/cud...
#graphics #cpp #rendering #raytracing
#cuda #cpp #graphics #raytracing #rendering
#cuda #cpp #graphics #raytracing #rendering
Check it out: github.com/RobinLmn/cuda-raytracer
#cuda #raytracing #cpp #graphics #rendering
Check it out: github.com/RobinLmn/cuda-raytracer
#cuda #raytracing #cpp #graphics #rendering