{"id":928,"date":"2019-08-15T07:15:08","date_gmt":"2019-08-15T07:15:08","guid":{"rendered":"http:\/\/janbosch.com\/blog\/?p=928"},"modified":"2019-08-15T07:15:10","modified_gmt":"2019-08-15T07:15:10","slug":"ai-the-fabric-of-computing","status":"publish","type":"post","link":"https:\/\/janbosch.com\/blog\/index.php\/2019\/08\/15\/ai-the-fabric-of-computing\/","title":{"rendered":"AI: the fabric of computing"},"content":{"rendered":"\n
\"\"
Image by Gerd Altmann from Pixabay<\/figcaption><\/figure>\n\n\n\n

During the summer, I spent a week at a summer school on deep learning (DL). There were several reasons to attend, but one was to simply learn more about this trending topic. In many ways, it was a wonderful, though humbling, experience as the field is progressing at a rate that\u2019s simply phenomenal. There are constant breakthroughs in the area and the quality of models, in terms of accuracy and loss, is still improving rapidly with every year significant improvements being reported by various research groups (especially the industrial ones).<\/p>\n\n\n\n

The \nability of the prototypes presented during the sessions was amazing. For\n instance, machine learning (ML) now exceeds the ability of theoretical \nphysicists to classify data in high energy physics. Deep learning models\n generate images that are amazingly realistic to the point of being \nindistinguishable from real photos. Classification in a wide range of \nproblem areas is exceeding human ability, such as in medical images and \nunderstanding human speech.<\/p>\n\n\n\n

Although machine learning \ntraditionally was applied on data that initially was intended for human \nconsumption, researchers and engineers are starting to realize that data\n at that level already has been processed and abstracted quite \nsignificantly. Consequently, there\u2019s a tendency to move machine learning\n closer to the lower levels in systems where the raw data is generated. \nThis allows ML techniques to detect features in the data that have been \nfiltered out of higher-level data that has been processed for human \nconsumption.<\/p>\n\n\n\n

The consequence is, of course, that we\u2019re entering an\n interesting paradox. On the one hand, ML techniques are implemented \ncloser to the places where raw data is generated \u2013 typically, sensors \nand devices that are deployed in the real world. These systems often \nhave quite limited computational resources. On the other hand, the \ncloser we are to the raw data, the larger the quantity of data actually \nis in practice. This requires careful balancing of computational \nresources and generated value \u2013 a trend that will fuel the continuous \ndevelopment of more powerful and more energy-efficient computation.<\/p>\n\n\n\n

Nevertheless,\n the trend is clear: AI and specifically ML and DL will be embedded in \nthe fabric of computing. Everywhere where data is generated or \nprocessed, ML\/DL algorithms will evaluate whether there are ways to \nimprove the outcome of system operation by classification, prediction \nand generation.<\/p>\n\n\n\n

Of course, today, we\u2019re far from this reality but \nduring this summer school, it became clear to me that the world clearly \nis moving in this direction. The challenge is what companies will \ncontrol this new technology space. In some areas, such as natural \nlanguage processing, the most advanced algorithms require such enormous \namounts of computational resources that only the FANGs of this world, ie\n companies like Facebook, Apple, Amazon, Netflix and Google, will be \nable to use the techniques. Similarly, it\u2019s clear that in machine \nlearning, larger data sets almost always lead to superior outcomes. In \none paper that was discussed during the summer school, the worst \nalgorithm with ten times the data performed better than the best \nalgorithm with one time the data. This means that there\u2019s a positive \nfeedback cycle between the size of the company in terms of customers and\n the quality of solutions it\u2019s able to put on the market as it\u2019s \ntypically the customers that generate the data that\u2019s used for training \npurposes.<\/p>\n\n\n\n

Although\n I\u2019m a happy customer of most of the FANGs, I do think that especially \nwhen it comes to AI\/ML\/DL, we should aim for democratization where \ncompanies of all sizes, as well as individuals, enjoy a level playing \nfield \u2013 hence my involvement with Peltarion. Embedded systems companies \nneed to have an explicit strategy to control their data, hypothesize the\n most promising use cases, experiment with different ML\/DL algorithms \nand find suitable ways to serve their customers better through data and \nAI.<\/p>\n\n\n\n

It sounds like a platitude but the fact is that data is the new oil. Software in general and ML\/DL in particular constitute the technology that refines the data and generates real, tangible value from this data. We can\u2019t afford to only have the FANGs of this world exploit this capability and I\u2019m glad that companies such as Peltarion<\/a> are looking to democratize the use of AI. However, it doesn\u2019t free you from the obligation to figure out how your company will be using data and AI to the advantage of your customers.<\/p>\n\n\n\n

To get more insights earlier, sign up for my newsletter at<\/em>jan@janbosch.com<\/em><\/a> or follow me on<\/em> janbosch.com\/blog<\/em><\/a>, LinkedIn (<\/em>linkedin.com\/in\/janbosch<\/em><\/a>) or Twitter (<\/em>@JanBosch<\/em><\/a>).<\/em><\/p>\n","protected":false},"excerpt":{"rendered":"

During the summer, I spent a week at a summer school on deep learning (DL). There were several reasons to attend, but one was to simply learn more about this trending topic. In many ways, it was a wonderful, though humbling, experience as the field is progressing at a rate that\u2019s simply phenomenal. There are … Read more<\/a><\/p>\n","protected":false},"author":2,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"generate_page_header":"","footnotes":""},"categories":[15,8,10],"tags":[],"_links":{"self":[{"href":"https:\/\/janbosch.com\/blog\/index.php\/wp-json\/wp\/v2\/posts\/928"}],"collection":[{"href":"https:\/\/janbosch.com\/blog\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/janbosch.com\/blog\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/janbosch.com\/blog\/index.php\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/janbosch.com\/blog\/index.php\/wp-json\/wp\/v2\/comments?post=928"}],"version-history":[{"count":2,"href":"https:\/\/janbosch.com\/blog\/index.php\/wp-json\/wp\/v2\/posts\/928\/revisions"}],"predecessor-version":[{"id":931,"href":"https:\/\/janbosch.com\/blog\/index.php\/wp-json\/wp\/v2\/posts\/928\/revisions\/931"}],"wp:attachment":[{"href":"https:\/\/janbosch.com\/blog\/index.php\/wp-json\/wp\/v2\/media?parent=928"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/janbosch.com\/blog\/index.php\/wp-json\/wp\/v2\/categories?post=928"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/janbosch.com\/blog\/index.php\/wp-json\/wp\/v2\/tags?post=928"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}