Artificial Intelligence
What AI Does Well
Reasoning: Legal assessment; financial asset management; financial application processing; games; autonomous weapons systems.
Knowledge: Medical diagnosis; druge creation; media recommendation; purchase prediction; financial market trading; fraud prevention.
Planning: Logistics; scheduling; navigation; physical and digital network optimization; predictive maintenance; demand forecasting; inventory management.
Communication: Voice control; intelligent agents, assistants and customer support; real-time translation of written and spoken languages; real-time transcription.
Perception: Autonomous vehicles; medical diagnosisl surveillance.
Machine Learning
The GOAL of most machine learning is to develop a prediction engine for a particular use case.
The defining characteristic of machine learning algorithms is that the quality of their predictions improve with experience.
Machine Learning Algorithms beyond Deep Learning
random forests: that create multitudes of detection trees to optimise a prediction
Bayesian networks: that use a probabilistic approach to analyse variables and the relationships between them
support vector machines: that are fed categorised examples and create models to assign new inputs to one of the categories
linear / logistic regression
boosted decision trees
note Each approach has its pros and cons, and combinations may be used (an "ensemble" approach).
Deep Learning
Why Deep Learning
We cannot specify the features to optimise ina way that is practical and reliable.
It avoids the programmer having to undertake the tasks of feature specification (defining the features to analyse from the data), or optimization (how the weigh the data to deliver an accurate prediction) -- the algorithm does both.
What Deep Learning Does Well
- recognise elements in pictures
- translate bewtween languages in real-time
- use speech to control devices
- predict how genetic variation will effect DNA transcription
- analyse sentiment in customer reviews
- detect tumours in medical images
What I am Interested in
keywords: enhanced targeting, intelligent job matching, automated assessment, friend suggestion, ad targeting, user interest prediction, supply / demand simulation, search result ranking