Summary:<\/em><\/strong>\u00a0 AI\/ML itself is the next big thing for many fields if you\u2019re on the outside looking in.\u00a0 But if you\u2019re a data scientist it\u2019s possible to see those advancements that will propel AI\/ML to its next phase of utility.<\/em><\/p>\n \u00a0<\/p>\n It\u2019s tempting to say that writers starting an article in this way should be flogged \u2026except that yours truly did recently start one with \u201cthe next most IMPORTANT thing in AI\/ML<\/u><\/em><\/a>\u2026\u201d\u00a0 Well that\u2019s clearly different isn\u2019t it \u2013 almost.<\/p>\n If you label something \u2018next big thing\u2019 it\u2019s evident you have a strong opinion \u2013 or your marketing department has no imagination.\u00a0<\/p>\n First of all, if you\u2019re on the outside of AI\/ML looking in, AI\/ML clearly is the next big thing.\u00a0 Most next-big-thing articles are actually in this category, explaining how AI\/ML can enhance everything from your dating life to your investment portfolio.<\/p>\n But if you\u2019re fortunate enough to be on the inside as our readers are then you know that the future of AI\/ML is developing along many different paths and some of those should be more important than others.\u00a0 Some are technical, some are applications, and some are even social or philosophical.\u00a0 So how to tell what the next big thing is or at least what the rankings should be.<\/p>\n To give this a little structure we need some guardrails.\u00a0 If it\u2019s \u2018next\u2019 then the time horizon needs to be reasonably short and the technical details need to be currently attainable.\u00a0 This leaves out all those Quora answers about transhumanism, AGI, nanorobotics, genetic predictions, and neuromorphic hardware.<\/p>\n Also if it\u2019s \u2018next\u2019 that means it\u2019s not yet fully realized.\u00a0 Alexa, facial recognition on my iPhone, my Roomba, and the chatbot at my bank are all wonderful, but those AI\/ML application technologies are fully realized.\u00a0 Been there, done that.<\/p>\n So after spending several hours consulting with the oracle of Google I would offer that there is an answer to this question but it depends on where you are standing when you look into the future.\u00a0 Here are four ways of looking at this that give somewhat different answers.<\/p>\n \u00a0<\/p>\n Things that lead to the greatest good for the greatest number of people<\/strong><\/span><\/p>\n This is really the question of what applications we have not yet cracked that bring the most good to the most people.\u00a0 The answer is healthcare both in the way you experience it currently in your doctor\u2019s office and what will shortly be possible applying AI\/ML to personalized medicine and drug discovery.<\/p>\n Despite all the hype about AI\/ML particularly in imaging being able to discern things in MRIs and lab tests that clinicians can\u2019t spot, you might be surprised to know that only about 1% of the roughly 6,000 US hospitals<\/u><\/em><\/a> have data science programs.<\/p>\n The causes of slow adoption are both financial and cultural with culture probably being the more important.\u00a0 Evidence is that AI\/ML developers are not paying close enough attention to how they are disrupting the workflow of doctors and clinicians that results in slow adoption.<\/p>\n Yes opportunities abound in:<\/p>\n Of all the AI\/ML opportunities in healthcare this one is actually furthest along largely because it\u2019s big pharma that pays, not the insurers.<\/p>\n The operational world of the clinician may be unique but at a business level hospitals and healthcare organizations share some marked similarities with the commercial world.<\/p>\n Determining whether and when a patient gets to see a doctor is a major control point for constraining healthcare costs, especially in single payer systems.\u00a0 The initial consult with the doctor is also a major time sink that could be at least partially automated.<\/p>\n If you want AI\/ML to succeed in improving healthcare it needs to get into the space between the doctor and the patient.\u00a0 Several major subsets of this opportunity need to be understood separately.<\/p>\n 4.1 Automated \/ Semi-automated interpretation of medical images.<\/p>\n 4.2 Enabling more accurate identification of disease subtypes \u2013 precision medicine.<\/p>\n 4.3 AI\/ML driven triage and prevention including developing protocols that give better outcomes, prevent bad outcomes, and that focus on anomalies and preventable harm.<\/p>\n \u00a0<\/p>\n Things that lead to cost savings, greater efficiency, and relieving mankind of low value repetitive work<\/strong><\/span><\/p>\n You may immediately think I mean robots and I do, but probably not in the way you\u2019re imagining.\u00a0 This is actually about Robotic (or Digital) Process Automation (RPA).<\/p>\n RPA as a rules-driven, non-AI\/ML technique has actually been around for some time.\u00a0 My feathers get ruffled when researchers want to say that current iterations of RPA are in fact an implementation of AI\/ML.<\/p>\n This can be really misleading.\u00a0 Based on a McKinsey study we reported that 47% of companies had at least one AI\/ML implementation in place<\/u><\/em><\/a>.\u00a0 Looking back at the data and the dominance of RPA as the most widely reported instance makes us think that the number is probably significantly over stated, maybe by as much as half.<\/p>\n But as RPA progresses it does increasingly have elements of embedded AI, just not the fancy or customized versions we data scientists are used to thinking about.\u00a0 For example, RPA platforms are increasingly embedding chatbots for both text and voice communication.\u00a0 Chatbots based on deep learning are a fully mature technology.<\/p>\n Also in there, increasingly you\u2019ll find image recognition tech based on CNNs.\u00a0 It\u2019s hidden in the document or screen recognition features.\u00a0 For example, when you direct your RPA engine to schedule an appointment on someone else\u2019s calendar or make an airline reservation it\u2019s faced with recognizing active elements on unknown screens that can be as simple as identifying what button to press to enter a command.\u00a0 Embedded AI\/ML computer vision tech can do this.<\/p>\n Most automated processes today are more in the nature of tasks rather than large scale processes as we think of them.\u00a0 That capability will continue to expand and if the AI\/ML is completely integrated and essentially hidden from the user, it still is, after all, AI\/ML.<\/p>\n \u00a0<\/p>\n Things that will move forward the technical capabilities of AI\/ML the most and the fastest.<\/strong><\/span><\/p>\n Machine learning hasn\u2019t had a breakthrough technique since about 2016, and AI deep neural nets have also been reduced to incremental improvements since then.\u00a0 I have two candidates in this column:<\/p>\n DRL is the current mashup of reinforcement learning with deep neural nets.\u00a0 The magic of reinforcement learning is that it can start with essentially a blank slate and develop better than human performance, especially since it does not require training data in the classical sense.\u00a0 In essence, reinforcement learning is a brute force iterative technique not much different from the evolutionary algorithms of ML.\u00a0 The addition of discriminators based on deep neural nets promises a breakthrough in performance.<\/p>\n The scope of what we might call awareness in our chatbots is quite limited.\u00a0 Ask it for the weather, the game score, or to play your favorite song and its advanced information retrieval algos will comply.\u00a0 But if we want Alexa to string together things it already knows about us to make helpful suggestions in our lives, that doesn\u2019t yet exist.<\/p>\n For example, why shouldn\u2019t Alexa remember that it\u2019s my mother\u2019s birthday and suggest a gift based on its knowledge of her likes and my interactions with her.\u00a0 That sort of stringing together the contextual elements of a \u2018story\u2019 will make our AI based virtual assistants much more valuable.<\/p>\n \u00a0<\/p>\n Things that mitigate resistance to AI\/ML so that society can readily accept this new capability<\/strong><\/span><\/p>\n How much we are bothered by or resistant to new AI\/ML tech seems to be based on how you ask the question.<\/p>\n If you ask the public if they like the convenience of internet shopping, the recommendations that are tailored specifically to them, or the actual reduction in cost that AI\/ML enabled targeting has created their response is overwhelmingly positive<\/u><\/em><\/a>.<\/p>\n Ask the same folks about privacy or let them hear about how the evil technology is biased against some ethnic or socio-economic group and their attitude changes dramatically.\u00a0 The media was quick to learn that publishing a thumb-sucker about a data breach gets our clicks faster than a funny cat video.<\/p>\n Although our most accurate models reach their classifications with a level of complexity that makes them impossible to explain, we are in fact having major breakthroughs in techniques for transparency and explainability.<\/p>\n Beyond simple transparency however, I would argue that our ability to show and explain causality is the next close-in frontier.\u00a0 The techniques are available but to this point there hasn\u2019t been enough demand to add this level of complexity to our algos.\u00a0 Now that the societal pressure has become great, causality may be the next most important thing<\/u><\/em><\/a> in allowing the public to embrace the benefits of AI\/ML.<\/p>\n \u00a0<\/p>\n \u00a0<\/p>\n Other articles by Bill Vorhies<\/u><\/em><\/a><\/p>\n \u00a0<\/p>\n About the author:\u00a0 Bill is Contributing Editor for Data Science Central.\u00a0 Bill is also President & Chief Data Scientist at Data-Magnum and has practiced as a data scientist since 2001.\u00a0 His articles have been read more than 2 million times.<\/p>\n He can be reached at:<\/p>\n Bill@DataScienceCentral.com<\/a> or<\/span> Bill@Data-Magnum.com<\/a><\/p>\n<\/div>\n Go to Source<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"<\/a>\u201cThe Next Big Thing in AI\/ML is\u2026\u201d as the lead to an article is probably the most overused trope since \u201conce upon a time\u201d.\u00a0 Seriously, just how many \u2018next big things\u2019 can there be?\u00a0 Is your incredulity not stretched every time you read that?<\/p>\n
\n
\n
\n
\n
\n
\n
\n