{"id":297,"date":"2017-05-01T08:24:51","date_gmt":"2017-05-01T08:24:51","guid":{"rendered":"http:\/\/buklijas.info\/blog\/?p=297"},"modified":"2017-02-19T15:24:28","modified_gmt":"2017-02-19T15:24:28","slug":"how-machines-learn-an-illustrated-guide-to-machine-learning-book-review","status":"publish","type":"post","link":"http:\/\/buklijas.info\/blog\/2017\/05\/01\/how-machines-learn-an-illustrated-guide-to-machine-learning-book-review\/","title":{"rendered":"How Machines Learn: An Illustrated Guide to Machine Learning, book review"},"content":{"rendered":"
<\/p>\n
Published on:<\/strong> 01.04.2017<\/p>\n Number of pages: 63 Conclusion<\/strong> Review<\/strong> This is my understanding of how machine learning is BIG<\/strong> change in software. Till last few yeas\/decades, all computer programs (all work that was done by computers) had two important characteristics: I maybe wrong, but AFAIK with machine learning, this is not true anymore.<\/p>\n Human was one who was designing and implementing algorithm. But, for example, with neural networks and their weight there is no way to understand why some input have particular weight. Argument is: computer can calculate average of 1000 number in less the second. In this argument I am exempting robots that work in conditions that are harmful for human, like high temperature, radiation, etc. In unsupervised learning goal is to find patterns that humans do not know about. I do not see big problem when neural network is used to do OCR, eg. you write some numbers on you touch screen, or when it classifies pictures of dogs and cats. But what if neural networks starts to be used for much more important decision ?<\/p>\n For example: visa acceptance, medical diagnostics, bank credit approval, decision if person is terrorist or not.<\/p>\n I would not be surprise that in future, there will be neural network making decisions, without anybody understanding what is the reason behind decision.<\/p>\n How can you even debug neural network when you have no idea how it decided on its input weights ?<\/p>\n How to know is it false negative or false positive ?<\/p>\n It will be interesting to see where machine learning will be in 10 years.<\/p>\n","protected":false},"excerpt":{"rendered":" Published on: 01.04.2017 Number of pages: 63 Written by: Helen Edwards Publish by: Self-publish Conclusion Best book for high-level understanding how today (in 2016\/2017) machine learning is working. Review Illustrations in book are very helpful. Without them it is much harder to understand machine learning concepts. This book covers all topics in machine learning, it […]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"spay_email":"","jetpack_publicize_message":"How Machines Learn: An Illustrated Guide to Machine Learning, #bookReview #machineLearning","jetpack_is_tweetstorm":false},"categories":[2],"tags":[6,18],"jetpack_featured_media_url":"","jetpack_publicize_connections":[],"yoast_head":"\n
\nWritten by: Helen Edwards<\/a>
\nPublish by: Self-publish<\/a><\/p>\n
\nBest book for high-level understanding how today (in 2016\/2017) machine learning is working.<\/p>\n
\nIllustrations in book are very helpful.
\nWithout them it is much harder to understand machine learning concepts.
\nThis book covers all topics in machine learning, it is not specializing on one aspect.<\/p>\nMore thoughts on machine learning<\/h4>\n
\nI would even say that machine learning is biggest change in software so far, it is a game changer.<\/p>\n
\n – Humans where making exact rules (algorithms) how computer will make decisions
\n – Everything that computer did, human could do it also<\/p>\nHumans where making exact rules (algorithms) how computer will make decisions<\/h6>\n
\nIf there was some BUG, person could understand what was the reason for the BUG.<\/p>\n
\nIt is not possible to understand “computer” reasoning behind it.
\nThis is BIG change in how algorithms work.
\nBecause computer is fine tuning algorithm, a human can not understand how or why in details.<\/p>\nEverything that computer did, human could do it also<\/h6>\n
\nHuman can do it also, but it need more time, much more time.
\nSo computer are used for this kind of work because it is more profitable, but still human can do it.<\/p>\n
\nI am just talking about software\/calculations.<\/p>\n
\nHumans can find some patterns across few dimensions, but even after 10 dimensions, what to speak of 100 or more it is not possible for human.<\/p>\nWhat will be result I do not know, but problems will arise<\/h6>\n
\nIf there is some error human can recognize error and cost of error is not important.
\nNot important in contexts, that result of error do not have huge impact on person life.
\nYou got number 5 instead of 3, or it showed you a dog instead of cat.<\/p>\n