Embed Size (px)
Citation preview
E X P E R I E N C E M O R E I N D U S T R Y N E W S
HOW IT ALL STARTED?
1952 A. Samuel wrote the firstcomputer-learning program – game of checkers
www.future-processing.com
1990Shift fromknowledge-drivenMachine Learning
to data-driven approach
2006 The very term Machine Learning comes into existence
2015Microsoft Kinectis able to track 20 human
features at a rateof 30 times per second
MACHINE LEARNI NGThe ability of a machine to study the environment in order to learn patterns and adjust its functions to it.
HOW CAN WE USE MACHINE LEARNING:
Analysis of human behaviour
Targeted advertising
Security Better diagnosis
of diseases
DEEP LEARNINGlayers of artificial neural networks
B AYES IAN NETW ORKS correlation between things
DEC IS ION TREE LEARNING observation of objects’ features
MACHINES CAN LEARN THANKS TO:
UNSUPERVISED MACHINE LEARNING
Lorem ipsum dolor sit
amet, consectetur
adipiscing elit. Sed
sollicitudin odio purus,
vel ultrices nibh mollis
quis. Donec convallis
vestibulum leo nec
malesuada. Sed eget
sapien vitae felis ornare
aliquam. Sed in nisl
malesuada, vulputate leo
ac, tristique augue.
Nulla in rhoncus urna.
Integer ultricies, augue
ut ultrices efficitur,
Lorem ipsum dolor sit
amet, consectetur
adipiscing elit. Sed
sollicitudin odio purus,
vel ultrices nibh
mollis quis. Donec
convallis vestibulum leo
nec malesuada. Sed eget
sapien vitae felis
ornare aliquam. Sed in
nisl malesuada,
vulputate leo ac,
tristique augue. Nulla
in rhoncus urna.
Integer ultricies, augue
ut ultrices efficitur,
purus elit imperdiet
nunc, at euismod orci
felis eu erat. Nulla
tellus turpis, elementum
sed suscipit at,
gravida non lacus. Sed
placerat nunc at
Lorem ipsum dolor sit amet, consectetur
adipiscing elit. Sed sollicitudin odio purus,
vel ultrices nibh mollis quis. Donec
convallis vestibulum leo nec malesuada. Sed
eget sapien vitae felis ornare aliquam. Sed
in nisl malesuada, vulputate leo ac,
tristique augue. Nulla in rhoncus urna.
Integer ultricies, augue ut ultrices
efficitur, purus elit imperdiet nunc, at
euismod orci felis eu erat. Nulla tellus
turpis, elementum sed suscipit at, gravida
non lacus. Sed placerat nunc at ullamcorper
euismod. Duis et suscipit metus, ac
venenatis tortor. Morbi sit amet ornare
orci. Vivamus sed nisl et nibh vulputate
scelerisque in id nibh. Donec in venenatis
dui. Aenean eget arcu consectetur, malesuada
magna pharetra, tristique ex. Duis tortor
metus, ultrices sed tempus et, mattis id
metus. Etiam nec erat magna. Donec elementum
Lorem ipsum dolor sit amet, consectetur
adipiscing elit. Sed sollicitudin odio
purus, vel ultrices nibh mollis quis.
Donec convallis vestibulum leo nec
malesuada. Sed eget sapien vitae felis
ornare aliquam. Sed in nisl malesuada,
vulputate leo ac, tristique augue. Nulla
in rhoncus urna. Integer ultricies, augue
ut ultrices efficitur, purus elit imperdiet
nunc, at euismod orci felis eu erat.
Nulla tellus turpis, elementum sed suscipit
at, gravida non lacus. Sed placerat nunc
at ullamcorper euismod. Duis et suscipit
metus, ac venenatis tortor. Morbi sit amet
ornare orci. Vivamus sed nisl et nibh
vulputate scelerisque in id nibh.
venenatis dui. Aenean eget arcu
REINFORCEMENTS LEARNING
Machines are learning based on the outcome
of various actions
Teaching by giving machines large volumes
of data to eventually make them able to
separate things based on different
aspects
T YPES OF MACHINE LEARNIN G
Teaching by taggingimages or objectsto help machineslearn algorithms
SUPE RVISE D MACH INE L EARNING
UNSUPERVISED MACHINE LEARNING
SOURCES:
SEE M ORE
FORBES.COMMACHINELEARNINGMASTERY.COM
