Page 16 - AeM_August_2017

RESEARCH, ANALYSIS & TRENDS

How deep learning turned sci-fi-esque

predictive intelligence into reality

20 years ago, IBM’s supercomputer Deep Blue defeated of top human chess players.
world chess champion Gary Kasparov in a historical first
victory for artificial intelligence. Today, supercomputers By implementing AI, the IBM supercomputer was able
are smart enough to easily beat not only chess players, to identify things it has seen before, consider all possi-
but also succeed in similarly sophisticated games, like ble moves, predict human responses, and then decide
the 3000-year-old Chinese game of Go, and most re- on the best move. This wouldn’t be possible without
cently, poker challenges against multiple human pros. training it to look at large amounts of data and use al-
These wins against humans were results from artificial gorithms that gave it the ability to perform tasks without
intelligence. any human intervention. This process led to what is
now known as “machine learning.”
Self-learning takes from human learning -
But it doesn’t end there. It takes even more intelligent
then outperforms It.
systems to defeat human players in more complex
The goal of artificial intelligence (AI) is to make comput- games like Chinese Go and especially Poker – there
ers as smart, or even smarter than human beings, by are not only billions of options to foresee, but they re-
giving them human-like thinking and reasoning abilities. quire “feeling” or intuition. This is where the process of
However, there are also many other ways to achieve deep learning comes into play.
this.
Deep learning is a highly innovative branch of machine
Several years ago, Deep Blue was taught by using hand learning that closely imitates the work of the human
-written functions, copying the knowledge and wisdom brain in processing data and creating patterns of deci-
16 August 2017 - Search Engine Marketing & Optimization