Is the link still available? No seeders anymore ?
Don't get anything from it.
Thanks
New Draft of “Reinforcement Learning: An Introduction, Second Edition”
21–30 of 36 posts
Re: New Draft of “Reinforcement Learning: An Introduction, Second Edition”
#22Earlier quoted context omitted.
Hey, I hope this doesn't come across as rude, but it seems like you're not a native English writer. The correct grammar for your question would be: Does anyone have a mirror? edit: Most people like when natives correct their English.
OP is a cat. The grammar is correct.
Wait, shouldn't that be "Can I haz mirror?"?
Re: New Draft of “Reinforcement Learning: An Introduction, Second Edition”
#23just reading page 15, arg max = maximal ..., I think that global maximum or local maximum is better than maximal. I would like to read all the interesting fruit of RL in just one hour, can someone suggest a short book for someone with advanced maths skills? Thanks a lot to the authors the book seems to be really interesting. Edit: In page 25, an extended example: tic-tac-toe the rule to update the value of each state…
The book is hundreds of pages long, if he diverges to talk about everything in Chapter 1 it would be a mess.
The scenario you describe is if alhpa=1, and it would do poorly. Try thinking about games where the opponent doesn't play an optimal game. Try thinking of stochastic environments.
Re: New Draft of “Reinforcement Learning: An Introduction, Second Edition”
#24I did my thesis about reinforced learning. Unfortunately didn't work with that afterwards, but I really think it's one of the most interesting approaches to machine learning for real world application.
What was your thesis? And what are some areas do you see RL being applied (besides the somewhat contrived game solvers we've been seeing)?
Re: New Draft of “Reinforcement Learning: An Introduction, Second Edition”
#25just reading page 15, arg max = maximal ..., I think that global maximum or local maximum is better than maximal. I would like to read all the interesting fruit of RL in just one hour, can someone suggest a short book for someone with advanced maths skills? Thanks a lot to the authors the book seems to be really interesting. Edit: In page 25, an extended example: tic-tac-toe the rule to update the value of each state…
The book is hundreds of pages long, if he diverges to talk about everything in Chapter 1 it would be a mess. The scenario you describe is if alhpa=1, and it would do poorly. Try thinking about games where the opponent doesn't play an optimal game. Try thinking of stochastic environments.
What I suggest is to use the function:
if v(s')==1 then 1 else the usual rule.
Re: New Draft of “Reinforcement Learning: An Introduction, Second Edition”
#26Earlier quoted context omitted.
I think they are parallel and can be used together, like Deep Reinforcement Learning.
I think they are parallel and can be used together Yep: https://en.wikipedia.org/wiki/AlphaGo#Algorithm
That's a sequential use. They got the initial strategy with DL, and made it stronger with MCTS
Re: New Draft of “Reinforcement Learning: An Introduction, Second Edition”
#27Is the link still available? No seeders anymore ? Don't get anything from it. Thanks
Try now.
There is also information about the book on the author's page about it on his website:
Re: New Draft of “Reinforcement Learning: An Introduction, Second Edition”
#28Earlier quoted context omitted.
The book is hundreds of pages long, if he diverges to talk about everything in Chapter 1 it would be a mess. The scenario you describe is if alhpa=1, and it would do poorly. Try thinking about games where the opponent doesn't play an optimal game. Try thinking of stochastic environments.
What I suggest is to use the function: if v(s')==1 then 1 else the usual rule.
Lets pretend alpha = 1 on a win and alpha = 0.1 on a loss.
Imagine a scenario where you play a game and the opponent plays poorly and you win; you then try and repeat the same thing again, this time the opponent has learnt from their mistakes and beats you. You'll keep playing the same losing move significantly more times because it worked that one time.
I don't know why everyone wants to second-guess the first chapter of the standard textbook in this space with what seems like no experience even thinking about this topic...
Re: New Draft of “Reinforcement Learning: An Introduction, Second Edition”
#29just reading page 15, arg max = maximal ..., I think that global maximum or local maximum is better than maximal. I would like to read all the interesting fruit of RL in just one hour, can someone suggest a short book for someone with advanced maths skills? Thanks a lot to the authors the book seems to be really interesting. Edit: In page 25, an extended example: tic-tac-toe the rule to update the value of each state…
Short and mathematical: "Algorithms for Reinforcement Learning". PDF available: https://sites.ualberta.ca/~szepesva/RLBook.html
Re: New Draft of “Reinforcement Learning: An Introduction, Second Edition”
#30Earlier quoted context omitted.
What I suggest is to use the function: if v(s')==1 then 1 else the usual rule.
Lets pretend alpha = 1 on a win and alpha = 0.1 on a loss. Imagine a scenario where you play a game and the opponent plays poorly and you win; you then try and repeat the same thing again, this time the opponent has learnt from their mistakes and beats you. You'll keep playing the same losing move significantly more times because it worked that one time. I don't know why everyone wants to second-guess the first chapt…
When you lose the value of v' change and so change the value of v.