Главная
Study mode:
on
1
Intro
2
Why Unity @ GDC ML Summit
3
How are studios using DRL?
4
Most common use case
5
Reinforcement learning in a nutshell
6
Test new levels or content using RL
7
A few effective approaches
8
Domain randomization
9
Using demonstrations to guide RL
10
How do we mitigate cost?
11
Increasing sample throughput
12
Increasing sample efficiency
13
Using RL for testing - final thoughts
14
Carry Castle
15
Challenges
16
RL setup for Source of Madness
17
Structuring the proper rewards
18
Lessons Learned - Visualizations
19
Lessons Learned - Balancing Rewards
20
Lessons Learned - Handling of Actions
Description:
Explore how deep reinforcement and imitation learning can revolutionize playtesting and NPC creation in game development. Delve into Unity's Jeffrey Shih's 2020 GDC Virtual Talk, which covers the most common use cases, effective approaches like domain randomization and guided demonstrations, and strategies to mitigate costs. Gain insights into real-world applications through case studies such as Carry Castle and Source of Madness, learning valuable lessons on visualizations, reward balancing, and action handling. Discover the potential of machine learning to scale and enhance game testing and character development processes.

Successfully Use Deep Reinforcement Learning in Testing and NPC Development

GDC
Add to list
#Conference Talks #GDC (Game Developers Conference) #Programming #Game Development #Computer Science #Machine Learning #Reinforcement Learning #Deep Reinforcement Learning #Imitation Learning
1 of 20

Intro