A computer program is said to learn from experience with respect to a class of tasks and a performance measure
P, if its performance at tasks as measured by
P improves with experience beyond a baseline of accuracy defined by: "Guessing the most frequently occurring outcome."
3Blue1Brown shows how trained neural networks simulate human abilities: https://www.youtube.com/watch?v=aircAruvnKk
CGP Grey shows how the genetic algorithm can simulate human abilities: https://www.youtube.com/watch?v=R9OHn5ZF4Uo
Machine Learning is 90% just a matter of being good at programming. Machine learning principles have been well-known since the 1960's, the difference today is that now a budget computer with $2500 worth of NVidia GPU's gets you more floating point operations per second than ten million dollars worth of computers in 1995 to 2005. In the last 20 years, computers are now outperforming humans and their `1*10^18` signal-processing neurons at narrowly defined tasks. Now $1000 worth of computer can outperform the best humans at Chess/Go in 1997, Wordplay understanding such as Jeopardy in 2011, Strategy games like Starcraft in 2019, and soon image recognition tasks such as Piloting vehicles. Musk has his Level 5 autonomy, but it only works in a narrow way on a few roads in the perfect conditions it was designed to operate in.
The mathematical part of Machine Learning really boils down to Calculus 3 of multiple variables you tried to understand in college. To excel in these classes: make sure to really understand
Discrete and continuous mathematics,
Calculus I through III and get good at transforming the theory to working code. One concept in Calculus you must understand backwards and forward (assuming you want to understand neural networks) is the derivative and integral of multiple variables. That 'finding slope of a point of a tangent on a curve' is the operating principle behind machine learning algorithms: knowing which way to tease the multiple in the
y=mx+b, to reduce error. https://www.youtube.com/watch?v=EkZGBdY0vlg
Member for 2 years, 2 months
8 profile views
Last seen Apr 5 at 18:01