Clean, labeled data requires data analysts with a combination of domain knowledge, and infrastructure engineers who can design and maintain robust data processing platforms. Looking toward narrow AI systems, much of the recent excitement involves systems that combine deep learning with additional techniques (reinforcement learning, probabilistic computing) and components (memory, knowledge, reasoning, and planning). Many current systems based on deep learning require big compute, big data, and big models. While researchers are seeking to build tools that are less dependent on large-scale pattern recognition, companies wanting to use deep learning as a machine learning technique can get started using tools that integrate with their existing big data platforms.
We'll be live streaming both the events on YouTube, so if you aren't able to make it, do watch the live streams (YouTube lets you set a reminder): The Fifth Elephant and Anthill Inside expose you to trends in data science, deep learning and artificial intelligence. Let's walk through the schedules for both events: At this point, we'd like to make a special mention about our diversity sponsor -- Intuit India -- for sponsoring child care facilities at The Fifth Elephant, and Anthill Inside. We'd like to talk about community for women and non-binary gender data scientists, the problems we are solving in the field, and how we can foster more diversity in data science. On that note, Intel has created a developer portal for ML engineers, data scientists and students with resources on optimized frameworks, and training for artificial intelligence, machine learning, and deep learning.
This summer, Dartmouth College's Neukom Institute for Computational Science held its annual Turing Tests in the Creative Arts. A system developed by Thomson Reuters Research Scientist Charese Smiley and Senior Software Engineer Hiroko Bretz took first prize in the poetry contest by creating a sonnet that judges thought most likely to be written by a human. Below is the sonnet created by Charese Smiley and Hiroko Bretz's software system: And be very careful crossing the streets. Our Cognitive Computing Center of Excellence focuses on exploring the rapidly developing field of cognitive computing and machine intelligence.
AI, Neural Networks, Machine Learning and other buzzwords are not new; they are with us from late 50s, but why did they become so much of a trend only now? The business focus changed from investing into so-called "artificial intelligence" to development of systems that could work with already gathered data, process and re-structurize it. Bayes was widely used in anti-spam, Markov's chains predicted criminal structure behavior, search engines developed decision trees to predict user input, speech and image recognition was no miracle anymore, and it was good. Basically, we returned to 50s -- we are trying to create universal structures, mimic human brain, and create entities that can process mixed data as our brains do.
Since the dawn of computing technology, developers created programs and algorithms by writing code that machines translate into precise instructions. Instead of code-writing the way a program solves a problem, the program "learns" to solve it on its own. In a not too dissimilar way than the human brain, unsupervised AI would recognize new patterns, label them on its own and classify them without human prior input. Per MIT Technology Review, Quoc Le (one of Google's Brain research scientists) has identified "unsupervised learning" as the biggest challenge to developing true AI that can learn without the need for labeled data.
The algorithms sifted through de-identified brain functional Magnetic Resonance Imaging (fMRI) data from an initiative called the Function Biomedical Informatics Research Network. The neuroimaging information used in this study was of 95 patients diagnosed with schizophrenia and schizoaffective disorders as well as individuals that served as a healthy control group. Essentially, the machine learning algorithms were able to explore these scans to create a model of schizophrenia that pinpoints brain connections most associated with the illness. The data also indicated that the diagnostic could distinguish between patients with schizophrenia and the control group with 74 percent accuracy, even as these images were collected from multiple sites through different means.
It follows a previous study that identified a genetic marker in dogs that sets them apart from wolves when it comes to human interaction, suggesting dogs developed a genetic condition through domestication that causes them to be so sociable. The findings challenge previous research that suggests dogs were domesticated twice by separate groups living in east and western Eurasia, instead revealing all modern dogs descended from animals that were domesticated by people living in Eurasia 20,000-40,000 years ago. In the study led by Princeton University biologist Bridgett von Holdt and researchers at Oregon State University, the team put 18 domesticated dogs and 10 captive human-socialized wolves to the test using problem-solving tasks. 'The genetic basis for the behavioural divergence between dogs and wolves has been poorly understood, especially with regard to dogs' success in human environments,' said Monique Udell, an animal scientist at Oregon State University.
One subset that has taken off is neural networks, systems that "learn" as humans do through training, turning experience into networks of simulated neurons. "A big problem is people treat AI or machine learning as being very neutral," said Tracy Chou, a software engineer who worked with machine learning at Pinterest Inc. "And a lot of that is people not understanding that it's humans who design these models and humans who choose the data they are trained on." It is a difficult enough problem to crack that the Defense Advanced Research Projects Agency, better known as Darpa, is funding researchers working on "explainable artificial intelligence." Here's why we're in this pickle: A good way to solve problems in computer science is for engineers to code a neural network--essentially a primitive brain--and train it by feeding it enormous piles of data.
Scientists at MIT are using Wi-Fi and AI to determine your emotional state. Without that tether, EQ Radio can't make assumptions about your heartbeat. The AI behind EQ Radio could figure out that you're stressed and cue the music without you even knowing you needed it. There's probably a pretty sizable market for parents as well – does your current router provide real-time EKG quality information about your sleeping newborn?
It was funded by the federal state of Baden-Württemberg's government which places great emphasis on research into intelligent systems: "With its Institute for Intelligent Systems in Tübingen and Stuttgart, the Max Planck Society has firmly established one of the key research fields in the digital transformation in Baden-Württemberg," indicated Minister-President Winfried Kretschmann. The new building accommodates the Tübingen section of the Max Planck Institute for Intelligent Systems which was only founded six years ago and has another Institute section in Stuttgart. The Institute has since become one of the world's leading research centres on intelligent systems and a wellspring for further activities supported by the federal state of Baden-Württemberg: "The mobility, medicine and mechanical engineering solutions of the future are inconceivable without artificial intelligence," pointed out Winfried Kretschmann. At the initiative of the Max Planck Society, the Max Planck Institute for Intelligent Systems, the federal state of Baden-Württemberg, the universities of Stuttgart and Tübingen and various industrial partners have joined forces in the "Cyber Valley" to step up research activities on intelligent systems – such as for autonomous vehicles – and to create a thriving environment for start-ups in this field.