Reinforcement Learning: Thompson Sampling & The Multi Armed Bandit Problem - Part 01

Dr. Soper discusses reinforcement learning in the context of Thompson Sampling and the famous Multi-Armed Bandit Problem. Topics include what the multi-armed bandit problem is, why the multi-armed bandit problem is important, what Thompson Sampling is, how Thompson Sampling works, and the role of the beta distribution in Thompson Sampling.
Previous lesson (Foundations of Reinforcement Learning): • Foundations of Reinfor...
Next lesson (Reinforcement Learning: Thompson Sampling & The Multi Armed Bandit Problem - Part 02): • Reinforcement Learning...

[@prabhudaskamath1353]

At last someone explaining in simple terms.. Thank you.

[@srinivasanbalan2469]

Thanks, Dr. Soper. You are awesome. Your voice is soothing.

[@rezamirabizadeh2215]

Thank you so so much Dr. Soper. Your content was very clear with exactly enough information to learn Thompson Sampling.

[@veramentegina]

Thank you.. This was very clear and articulate delivery of the subject.

[@laimeilin6708]

I finally understood the slot machine analogy haha, thanks so much Dr.Daniel Soper, look forward to more content from you x

[@carlosroquesuarezgurruchag8681]

Thank you for the time and the explanation. It was really clear !

[@shangqunyu5445]

thanks you Dr Soper!

[@akramsystems]

concise and clear beautifully done!

[@yoggi1222]

Excellent Explainations !

[@aixueer4ever]

Thank you very much. The shaded area at 14:16 is inaccurate. Only the left half of it means sampling from red distribution has bigger chances than blue.

[@JonathanWeins]

Great video!

[@beS.M.A.R.T]

excellent presentation

[@tldyesterday]

Thank you so much for this!!

[@hessamjamalkhah9781]

It was great, thank you

[@afraimgershenzon8014]

Well done

[@NurilGamer999]

wow this is good. thank you Sir

[@bhavnagupta3045]

This helped me alot thanks

[@ajiths1689]

well explained

[@sigmatau8231]

finally, an immediately digestible explanation; thank you.

[@PasinduTennageprofile]

The best!

[@antwidavid389]

The multi-armed bandit problem is like the basic economic problem of unlimited wants exceeding limited resources, which results in scarcity, and thus, an opportunity cost when making a decision.

[@pattiknuth4822]

A 5 minute lecture crammed into 16 minutes. If you want to know how to implement Thompson sampling you won't find it in this video.

[@sergiolenoo]

the principle of a good teacher is to make even those with difficulty understand what is being taught. Not everyone has prior knowledge of the subject, so, it's great that he explains it slowly.

[@amromustafa117]

very well explained