@amodm But, isn't the problem not being able to find that "wobbling plate" ?

@amodm This was amazing to read

@amodm You can't outwork someone who is in for the play and fun.

@amodm I wished I had read this before. Thanks for sharing Amod!

@amodm Sweeeeeet

@amodm It was effortless. It was easy to play with these things. I almost tried to resist it! There was no importance to what I was doing, but ultimately there was. The diagrams and the whole business that I got the Nobel Prize for came from that piddling around with the wobbling plate.

@amodm Loved it

@amodm Amazing, thanks for sharing!

@amodm This reminds me how side projects are fun but they can stop being fun if you work on them full time and force yourself to work.

@amodm Thanks a lot for sharing ! God knows how many people need to see this. This reminds of an old Mick Jagger interview: how they found thw same jamming when done at a recording studio for paid work was exhausting as against doing it in their garage. So they started doing a prejam

@amodm So good! In reinforcement learning the "playing vs accomplishing" is mapped as "exploration vs exploitation". During the exploration phase, the system learns, and improves itself, and in exploitation phase it only applies whatever it learnt (so it will get boring :P).