Hello Daniel, I just can't belive : Thanks to your series I'm working with python, jupyter and execute IA exercises and It's working!!! I studied mathematics and computer science 35 years ago, I just remember why I loved my carrer. Just the best course! (greetings from México)
Thank you so much! Could you please suggest a good reference book that focuses mainly on reinforcement learning? I would prefer that it starts from scratch.
Just found this video and is the best I've ever seen on this topic. What would be the procedure if the "conversion rate" is changing over time?, my guessing is to take only the last "n" data for the sampling (last 10, last 20, etc.) but in this case which is the "n" minimum value that can be used to apply in the beta distribution?
Also, when indexing outcomes we use [turn_index] so does it mean that the outcome depends on the number of turn we are in? I thought every spin is independent. Why is the index of the turn affecting the outcome?
Hello Daniel, I just can't belive : Thanks to your series I'm working with python, jupyter and execute IA exercises and It's working!!! I studied mathematics and computer science 35 years ago, I just remember why I loved my carrer. Just the best course! (greetings from México)
Excellent example and explanation, thank you
I hope you decide to continue your videos, they are just perfect
What an amazing video! You are the best!
Hello Daniel! Thank you so much for these videos! They are amazing and really helpful! You are a great one
Great video and explanation of Thompson Sampling and its practical application through the multi-armed bandit scenario.
Thank you for the clear explanation.
Thank you so much! Could you please suggest a good reference book that focuses mainly on reinforcement learning? I would prefer that it starts from scratch.
Just found this video and is the best I've ever seen on this topic. What would be the procedure if the "conversion rate" is changing over time?, my guessing is to take only the last "n" data for the sampling (last 10, last 20, etc.) but in this case which is the "n" minimum value that can be used to apply in the beta distribution?
Why if the random number is less than conversion rate , the outcome is 1? It should be other way round no?
thank you. you are the best!!
That is great! Thanksss you so much!!q
Nice music !
You identified that we have $1000 initially. However, we played 1000 turns and for each turn we used each (6) machine. So didnt we spent $6000?
Also, when indexing outcomes we use [turn_index] so does it mean that the outcome depends on the number of turn we are in? I thought every spin is independent. Why is the index of the turn affecting the outcome?