ideas.txt

ideas.txt

distribution data => distribution name, params supervised ANN classifier
eg. 
Poisson(5) data 1000 samples => ("poisson", 5)
Uniform(10) data 1000 samples => ("uniform", 10)
Normal(mu, sig) data 1000 samples => ("normal", mu, sig)
etc.

synthetic data - https://arxiv.org/abs/1804.06516
make lots of synthetic scenes train on them and use model on test. placing cars in arbitrary scenes. very similar to generating swish by randomly iterating on diff activations until one happens to work. generate millions of distributions by varying mu sigma until you learn gaussian ....

detection of workout videos based on content ( rhythmic beat (audio FFT), racy content, text tags by actor actress lyricist md etc )

simulate non-binom ( gaussian poisson etc) with coin toss