Discussion of anything and everything relating to chess playing software and machines.
Moderators: hgm, Dann Corbit, Harvey Williamson
Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
-
amchess
- Posts: 153
- Joined: Tue Dec 05, 2017 1:42 pm
Post
by amchess » Tue Sep 10, 2019 5:45 pm
zullil wrote: ↑Sat Sep 07, 2019 2:30 pm
Forgive me, but it seems all you've done is add a "permanent" hash table. This has been done many times before, even for Stockfish. How is this different? I mean, who would want to publish this?
We are using the entry level techniques in RL:Q learning.
The search and evaluation function provides the penalties and bonuses you want.
But we are not here to teach you reinforcent learning
-
zullil
- Posts: 6442
- Joined: Mon Jan 08, 2007 11:31 pm
- Location: PA USA
- Full name: Louis Zulli
Post
by zullil » Tue Sep 10, 2019 6:43 pm
amchess wrote: ↑Tue Sep 10, 2019 5:45 pm
zullil wrote: ↑Sat Sep 07, 2019 2:30 pm
Forgive me, but it seems all you've done is add a "permanent" hash table. This has been done many times before, even for Stockfish. How is this different? I mean, who would want to publish this?
We are using the entry level techniques in RL:Q learning.
The search and evaluation function provides the penalties and bonuses you want.
But we are not here to teach you reinforcent learning
Never asked for you to teach me anything. If nothing else, you might want to review your article/abstract, since there's absolutely nothing in it that seems to qualify as
reinforcement learning. What you present in the "article" seems to be nothing more than a hash table saved on disk, an idea that's been around for a long time. Perhaps if I review the source code I'll see something more... .
-
amchess
- Posts: 153
- Joined: Tue Dec 05, 2017 1:42 pm
Post
by amchess » Tue Sep 10, 2019 10:05 pm
You might as well review you understanding on q learning because it is the simplest algorithm in reinforcement learning. At first you wanted penalties and bonuses, we gave you. You can tell us what you want next and we shall give you. Well, it looks like your definition of reinforcement learning is questionable
-
zullil
- Posts: 6442
- Joined: Mon Jan 08, 2007 11:31 pm
- Location: PA USA
- Full name: Louis Zulli
Post
by zullil » Tue Sep 10, 2019 11:39 pm
amchess wrote: ↑Tue Sep 10, 2019 10:05 pm
You might as well review you understanding on q learning because it is the simplest algorithm in reinforcement learning. At first you wanted penalties and bonuses, we gave you. You can tell us what you want next and we shall give you. Well, it looks like your definition of reinforcement learning is questionable
I never mentioned bonuses or penalties, so you must have me confused with someone else. Good luck with your project.
-
amchess
- Posts: 153
- Joined: Tue Dec 05, 2017 1:42 pm
Post
by amchess » Fri Sep 20, 2019 8:49 pm
Builds for gcc 8.
Bug corrected.
Now, the file is written only when the gui sends the quit command.
Important
An infinite analysis must be stopped before any other operation.
If not, the learning HashTable is not filled and the experience.bin file
is incorrectly written.
https://github.com/amchess/BrainLearn/r ... /tag/4.2.1
-
Leo
- Posts: 993
- Joined: Fri Sep 16, 2016 4:55 pm
- Location: USA/Minnesota
- Full name: Leo Anger
Post
by Leo » Sun Jan 12, 2020 4:53 pm
How do I obtain the private offline learning application?
Advanced Micro Devices fan.