Deep learning and computer chess (part 2)
A machine would need more than 10^90 years to make the first chess move using brute force method. To address this problem, various strategies based on minimax algorithms and deep learning advancements have surfaced throughout time. One innovation that is able to address this problem is giraffe ar...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project |
| Language: | English |
| Published: |
Nanyang Technological University
2022
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/162926 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Nanyang Technological University |
| Language: | English |
| Summary: | A machine would need more than 10^90 years to make the first chess move using brute force
method. To address this problem, various strategies based on minimax algorithms and deep
learning advancements have surfaced throughout time. One innovation that is able to address
this problem is giraffe architecture. The data set was carefully chosen from chess matches
between different grandmasters from around the world. To test the idea and see how the models
react, various Giraffe models with different parameter settings are created. |
|---|