{"id":2155,"date":"2019-12-24T03:11:24","date_gmt":"2019-12-24T00:11:24","guid":{"rendered":"http:\/\/kusuaks7\/?p=1760"},"modified":"2024-02-02T15:11:29","modified_gmt":"2024-02-02T15:11:29","slug":"architects-of-intelligence-a-reflection-on-the-now-and-future-of-ai","status":"publish","type":"post","link":"https:\/\/www.experfy.com\/blog\/ai-ml\/architects-of-intelligence-a-reflection-on-the-now-and-future-of-ai\/","title":{"rendered":"Architects of Intelligence: A reflection on the now and future of AI"},"content":{"rendered":"\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\tThe original vision of the pioneers of artificial intelligence was to create human-level AI, machines that could understand, reason, and act like humans. But six decades of research have proven that artificial general intelligence (AGI) is a very tough nut to crack.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\tThe field has seen tremendous progress, especially with an explosion of innovation in\u00a0deep learning<\/a>\u00a0and\u00a0neural networks<\/a>\u00a0in recent years. But we also face many fundamental questions: What is the path towards AGI? What are the capabilities and limits of current AI technologies? How do we know that we\u2019ve achieved AGI? And how far are we from human-level AI?\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\tWith all the hype and confusion surrounding AI, it\u2019s difficult to answer those questions. Our\u00a0AI models can beat human professionals<\/a>\u00a0at the most complicated games. But at the same time, they can\u2019t replicate some of the simplest cognitive functions of humans.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t\nAuthor and futurist Martin Ford has done a great job of answering these questions in his book\u00a0Architects of Intelligence: The Truth about AI from the People Building it<\/a><\/em>.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\tFord\u2019s book is a compilation of interviews with 23 leading AI scientists and experts. It discusses, among other things, the current state of AI and the path to artificial general intelligence.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\tArchitects of Intelligence\u00a0<\/em>answers many of the fundamental questions, and like everything AI, leaves us with many more.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t

Deep learning is here to stay<\/h2><\/h2>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\tWith deep learning being the cutting edge of AI,\u00a0scientists are divided<\/a>\u00a0over the extent of its capabilities and limits. But Ford\u2019s interviews show that AI experts agree that deep learning will be crucial to reach artificial general intelligence.\n\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t\u201cThe scientific concepts that are behind deep learning and the years of progress made in this field, means that for the most part, many of the concepts behind deep learning and neural networks are here to stay. Simply put, they are incredibly powerful. In fact, they are probably going to help us better understand how animal and human brains learn complex things,\u201d says Yoshua Bengio, computer science professor at the University of Montreal.\nArchitects of Intelligence: The truth about AI from the people building it<\/span><\/span><\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\"\"\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

Architects of Intelligence: The truth about AI from the people building it<\/span><\/span><\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t

\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\tBengio\u2019s remark is a kind of expectable, given that he is\u00a0one of the pioneers of deep learning<\/a>. But deep learning also gains appraise from its critics such as\u00a0neuroscientist and AI expert Gary Marcus<\/a>.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t\u201cI see deep learning as a useful tool for doing pattern classification, which is one problem that any intelligent agent needs to do. We should either keep it around for that, or replace it with something that does similar work more efficiently, which I do think is possible,\u201d Marcus says.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\tArchitects of Intelligence\u00a0<\/em>also discusses many fields that have benefited from advances in deep learning, including\u00a0computer vision<\/a>\u00a0and\u00a0natural language processing<\/a>.\n\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t

The challenges of deep learning<\/h2><\/h2>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\tMeanwhile, the interviewed scientists also acknowledge that current technologies have some hurdles to overcome if we want to achieve human-level AI.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t\u201cThe success today of neural networks and deep learning mostly involve supervised pattern recognition, which means that it\u2019s a very narrow sliver of capabilities compared to general human intelligence,\u201d says Fei-Fei Li, professor of computer science at Stanford and chief scientist at Google Cloud. Other scientists interviewed in\u00a0Architects of Intelligence\u00a0<\/em>echo those remarks, including Yann LeCun, another deep learning pioneer.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\tSupervised learning is the process of creating AI models by\u00a0training them on lots of labeled examples<\/a>. While supervised learning helps solve many problems in AI, it also poses some challenges. In many domains, labeled data is scarce or requires extensive human efforts.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\tMany of Ford\u2019s interviews discuss the challenges of current AI, including its\u00a0application to narrow domains<\/a>, its overreliance on data, and its\u00a0limited understanding<\/a>\u00a0of the meaning of language. His interview with Gary Marcus and Oren Etzioni, CEO of Allen Institute for Artificial Intelligence, dig deep into these challenges and what\u2019s preventing deep learning from solving problems that are easy to tackle for a human child.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t\n\u201cI think the reality is that deep learning and neural networks are particularly nice tools in our toolbox, but it\u2019s a tool that still leaves us with a number of problems like reasoning, background knowledge, common sense, and many others largely unsolved,\u201d Etzioni says.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t

Is hybrid AI the right path to human-level intelligence?<\/h2><\/h2>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\"\"\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\tSeveral of the experts Ford interviewed in\u00a0Architects of Intelligence<\/em>\u00a0believe that the combination of neural networks and classic, rule-based AI will help overcome the limits of deep learning.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t\u201cOn balance, there\u2019s been a shift from traditional tools toward deep learning, especially when you have a lot of data, but there are still plenty of problems in the world where you have only small datasets, and then the skill is in designing the hybrid and getting the right mix of techniques,\u201d says Andrew Ng, adjunct professor of computer science at Stanford University, co-founder of Google Brain, and former chief scientist at Baidu.\n\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t\n\u201cHumans have all kinds of common-sense reasoning, and that has to be part of the solution. It\u2019s not well captured by deep learning. In my view, we need to bring together symbol manipulation, which has a strong history in AI, with deep learning. They have been treated separately for too long, and it\u2019s time to bring them together,\u201d Marcus says.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\tAnd Joshua Tenenbaum, professor of computational cognitive science at MIT, posits that we must combine the achievements from symbolic AI, probabilistic and causal models, and neural networks to solve the challenges of deep learning.\n\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t\u201cEach of these ideas has had their rise and fall, with each one contributing something, but neural networks have really had their biggest successes in the last few years. I\u2019ve been interested in how we bring these ideas together. How do we combine the best of these ideas to build frameworks and languages for intelligent systems and for understanding human intelligence?\u201d Tenenbaum says.\n\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\tTenenbaum recently headed a team of researchers who developed the\u00a0Neuro-symbolic Concept Learner<\/a>. The NSCL is a hybrid AI model that combines neural nets and symbolic AI to solve problems. The results of the researchers\u2019 work show that NSCL can learn new tasks with much less data than pure neural network\u2013based models require. Hybrid AI models are also\u00a0explainable<\/a>\u00a0as opposed to being opaque black boxes.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\tBut not everyone is a fan of hybrid AI models.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t\u201cNote that your brain is all neural networks. We have to come up with different architectures and different training frameworks that can do the kinds of things that classical AI was trying to do, like reasoning, inferring an explanation for what you\u2019re seeing and planning,\u201d Bengio says.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\tGeoffrey Hinton, another deep learning pioneer, is also critical toward hybrid approaches. In his interview with Ford, he compares hybrid AI to combining electric motors and internal combustion engines. \u201cThat\u2019s how people in conventional AI are thinking. They have to admit that deep learning is doing amazing things, and they want to use deep learning as a kind of low-level servant to provide them with what they need to make their symbolic reasoning work,\u201d Hinton says. \u201cIt\u2019s just an attempt to hang on to the view they already have, without really comprehending that they\u2019re being swept away.\u201d\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t

How do we know we\u2019ve achieved general AI?<\/h2><\/h2>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\"\"\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\tSince the time of Alan Turing, the father of modern computer science, the \u201cimitation game,\u201d which later became known as the Turing Test, has been the principal benchmark of determining whether we\u2019ve developed \u201cthinking machines.\u201d The idea behind the Turing Test is that an AI\u2014say a chatbot\u2014must be able to fool humans into thinking it is a human.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t\nWhile there\u2019s a lot of debate over whether the Turing Test is a real measure of advances in AI, most scientists agree that language understanding is an essential part of any real intelligence system.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t\u201c[Deep-learning based natural-language systems] systems are really good at statistical learning, pattern recognition and large-scale data analysis, but they don\u2019t go below the surface,\u201d says Barbara Grosz, Higgins Professor of Natural Sciences at Harvard University. \u201cThey can\u2019t reason about the purposes behind what someone says. Put another way, they ignore the intentional structure component of dialogue. Deep-learning based systems more generally lack other hallmarks of intelligence: they cannot do counterfactual reasoning or common-sense reasoning.\u201d\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\tEtzioni says that one of the essential stepping stones toward artificial general intelligence would be to develop AI programs that can handle multiple tasks. \u201cAn AI program that\u2019s able to both do language and vision, it\u2019s able to play board games and cross the street, it\u2019s able to walk and chew gum. Yes, that is a joke, but I think it is important for AI to have the ability to do much more complex things,\u201d he says.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\tPresently, every task needs a separate AI, and efforts to create generalized AI models have had limited success.\n\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\tJames Manyika, the Chairman and Director of McKinsey Global Institute, makes a fun proposition. \u201cUntil you get a system that can enter an average and previously unknown American home and somehow figure out how to make a cup of coffee, we\u2019ve not solved AGI,\u201d he says.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\tWhile it might sound silly, Manyika\u2019s proposition is a very serious test of the general problem\u2013solving capabilities of AI. Marcus discusses why simple tasks such as Manyika\u2019s coffee challenge are\u00a0tough for current blends of AI<\/a>.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\tAnd of course, most scientists agree that real artificial intelligence shouldn\u2019t need so much labeled data to learn. \u201cIn order to get really highly effective machine intelligent systems, we also need algorithms that can make more use of unsupervised and unlabeled data. As humans, we tend to organize a lot of our world knowledge in causal terms, and that\u2019s something that is not really done much by current neural networks,\u201d says Nick Bostrom, professor at the University of Oxford and the write of New York Time bestseller\u00a0Superintelligence: Paths, Dangers, Strategies.<\/em>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\tIn the end,\u00a0Architects of Intelligence\u00a0<\/em>reminds us that we still have a lot to learn about intelligence, many more questions to answer, and some questions to discover.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\tTo quote Etzioni: \u201cPeople see these amazing achievements, like a program that beats people in Go and they say, \u2018Wow! Intelligence must be around the corner.\u2019 But when you get to these more nuanced things like natural language, or reasoning over knowledge, it turns out that we don\u2019t even know, in some sense, the right questions to ask.\u201d\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<\/div>\n\t\t","protected":false},"excerpt":{"rendered":"

The original vision of the pioneers of artificial intelligence was to create human-level AI, machines that could understand, reason, and act like humans. But six decades of research have proven that artificial general intelligence (AGI) is a very tough nut to crack.The field has seen tremendous progress, especially with an explosion of innovation in\u00a0deep learning\u00a0and\u00a0neural<\/p>\n","protected":false},"author":109,"featured_media":3143,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"content-type":"","footnotes":""},"categories":[183],"tags":[97],"ppma_author":[1946],"class_list":["post-2155","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-ai-ml","tag-artificial-intelligence"],"authors":[{"term_id":1946,"user_id":109,"is_guest":0,"slug":"ben-dickson","display_name":"Ben Dickson","avatar_url":"https:\/\/www.experfy.com\/blog\/wp-content\/uploads\/2020\/04\/medium_8aaf6bea-c4c1-455f-8156-8007d70910f8-150x150.jpg","user_url":"https:\/\/bdtechtalks.com\/","last_name":"Dickson","first_name":"Ben","job_title":"","description":"Ben Dickson is an experienced software engineer and tech blogger. He contributes regularly to major tech websites such as the Next Web, the Daily Dot, PCMag.com, Cointelegraph, VentureBeat, International Business Times UK, and The Huffington Post."}],"_links":{"self":[{"href":"https:\/\/www.experfy.com\/blog\/wp-json\/wp\/v2\/posts\/2155","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.experfy.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.experfy.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.experfy.com\/blog\/wp-json\/wp\/v2\/users\/109"}],"replies":[{"embeddable":true,"href":"https:\/\/www.experfy.com\/blog\/wp-json\/wp\/v2\/comments?post=2155"}],"version-history":[{"count":4,"href":"https:\/\/www.experfy.com\/blog\/wp-json\/wp\/v2\/posts\/2155\/revisions"}],"predecessor-version":[{"id":35844,"href":"https:\/\/www.experfy.com\/blog\/wp-json\/wp\/v2\/posts\/2155\/revisions\/35844"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.experfy.com\/blog\/wp-json\/wp\/v2\/media\/3143"}],"wp:attachment":[{"href":"https:\/\/www.experfy.com\/blog\/wp-json\/wp\/v2\/media?parent=2155"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.experfy.com\/blog\/wp-json\/wp\/v2\/categories?post=2155"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.experfy.com\/blog\/wp-json\/wp\/v2\/tags?post=2155"},{"taxonomy":"author","embeddable":true,"href":"https:\/\/www.experfy.com\/blog\/wp-json\/wp\/v2\/ppma_author?post=2155"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}