nus-cs3281 / 2019

0 stars 1 forks source link

Machine Learning #49

Open amrut-prabhu opened 5 years ago

amrut-prabhu commented 5 years ago

Machine Learning slides

PUNCH: Youube keeps recommending me videos before I even have to search for them. It learns from my past behaviour. This is possible due to Machine Learning (ML).

WIIFY: Since you are a CS student, someone is bound to ask you about ML, and you wouldn't want to respond by saying that you don’t know. I'll help you understand what ML is and what it’s capable of.

Key points:

  1. Machine Learning is easy to intuitively understand.
    • Uses data to get the best mathematical model to make predictions
    • Same basic process cycle for all ML models
  2. Machine Learning is not the future anymore, it is the present.
    • Widespread usage in real world applications
    • Increasing demand for ML-related jobs
  3. ML is good enough for many purposes, but not all.
    • Useful for tasks that require lot of data
    • Limited in capabilities and is not 100% accurate

Intended impact: