Eric Leschinski

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, how trained neural networks simulate human abilities: https://www.youtube.com/watch?v=aircAruvnKk

CGP Grey, how the genetic algorithm can simulate human abilities: https://www.youtube.com/watch?v=R9OHn5ZF4Uo

Machine Learning is mostly just a matter of getting good at programming. The annoying part of machine learning is data transformation from a given arrangement to a needed arrangement. To understand which kind of arrangement is needed, you must correctly reason about Calculus enough to know why the slope of a tangent of a point on a curve is a source of information gain.

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 (or cheap EC2 instance) gets you more 32-bit floating point operations per second than ten million dollars worth of computer in 1990. In the last 20 years, computers are outperforming humans (1*10^18 signal-processing neurons plus 100TB of bio-memory) at narrowly defined tasks. Today in 2020, a $2700 macbook can outperform the best humans at Chess/Go, do wordplay understanding such as Jeopardy, and soon image recognition tasks such as piloting vehicles will be something $2700 worth of computer can do better than 90% of humans. Musk has Level 5 autonomy in his flagship product (https://www.youtube.com/watch?v=tlThdr3O5Qo), but it only works in ideal ideal conditions. It works well enough, but he can't ship it to customers because millions of people will stop monitoring the vehicle and cognitively zone-out with hands on the wheel.

The mathematical part of Machine Learning boils down to "Calculus 3 of multiple variables" and "Being good at translating Math to code, and code to math with intuition around both". You remember studying y=mx+b in high school, that is the beginning of machine learning. To excel in machine learning: make sure to really understand all math themed classes: Algebra, Discrete and continuous mathematics, Probability, Statistics, Set/decision theory, Calculus I through III and get good at reading Math equations and computer code, and having an intuition for what it will do and why.

The reason machine learning is good for everyone is because it offloads real work from a human, who would have had to allocate neurons and bio-memory with the expenditure of glucose and time, onto a computer that is able to do the same work but at a tenth of the overall cost.


people reached
  • Member for 7 years, 3 months
  • 235 profile views
  • Last seen Jan 24 at 17:27

Top tags (32)

Score 80
Posts 4
Posts % 25
Score 65
Posts 3
Score 29
Posts 4
Score 18
Posts 2
Score 11
Posts 2
Score 7
Posts 2

Top posts (16) All Questions Answers | Votes Newest

View all questions and answers

Badges (34)