Dinesh Verma is an IBM Fellow, the company's pre-eminent technical distinction granted in recognition of outstanding and sustained technical achievements and leadership in engineering. Dinesh has worked in IBM Research for nearly 25 years, holds more than 150 patents, is a member of the IBM Academy of Technology, and heads a team that is focused on Distributed Artificial Intelligence (AI). The IBM THINK Blog caught up with Dinesh recently to talk about his current work, as well as his career at IBM. The following is an excerpt and is part of our Perspectives series featuring stories by and about IBMers who take the "long view." THINK: Can you tell us a little bit about your role at IBM? Dinesh Verma: I lead the Distributed AI team at IBM Research at the Thomas J. Watson Research Center in Yorktown, NY.

As AI continues to advance, what future jobs are just over the horizon? Despite fears of automation taking away jobs, the need for skilled humans to operate, utilise and advance technologies will remain unequivocally necessary. While there are plenty of people who fear robots taking over their jobs, there are also many important positives to automation. This starts with having robots in the workplace to treat like robots, and revaluing human employees as actual humans with a need for purpose and work-life balance. Automation, and augmented and virtual reality (AR/VR) all lead to the idea that human workers will be valued for their uniquely human skills, such as creativity and innovative thinking.

This book was designed around major data structures, operations, and techniques in linear algebra that are directly relevant to machine learning algorithms. There are a lot of things you could learn about linear algebra, from theory to abstract concepts to APIs. My goal is to take you straight to developing an intuition for the elements you must understand with laser-focused tutorials. I designed the tutorials to focus on how to get things done with linear algebra. They give you the tools to both rapidly understand and apply each technique or operation. Each tutorial is designed to take you about one hour to read through and complete, excluding the extensions and further reading. You can choose to work through the lessons one per day, one per week, or at your own pace. I think momentum is critically important, and this book is intended to be read and used, not to sit idle. I would recommend picking a schedule and sticking to it.

So you've taken the plunge. You want to become a data scientist. There are far too many resources out there. How do you decide the starting point? Did you miss out on topics you should have studied?

An intelligent robot agent based on domain ontology, machine learning mechanism, and Fuzzy Markup Language (FML) for students and robot co-learning is presented in this paper. The machine-human co-learning model is established to help various students learn the mathematical concepts based on their learning ability and performance. Meanwhile, the robot acts as a teacher's assistant to co-learn with children in the class. The FML-based knowledge base and rule base are embedded in the robot so that the teachers can get feedback from the robot on whether students make progress or not. Next, we inferred students' learning performance based on learning content's difficulty and students' ability, concentration level, as well as teamwork sprit in the class. Experimental results show that learning with the robot is helpful for disadvantaged and below-basic children. Moreover, the accuracy of the intelligent FML-based agent for student learning is increased after machine learning mechanism.

Model distillation was originally designed to distill knowledge from a large, complex teacher model to a faster, simpler student model without significant loss in prediction accuracy. We investigate model distillation for another goal -- transparency -- investigating if fully-connected neural networks can be distilled into models that are transparent or interpretable in some sense. Our teacher models are multilayer perceptrons, and we try two types of student models: (1) tree-based generalized additive models (GA2Ms), a type of boosted, short tree (2) gradient boosted trees (GBTs). More transparent student models are forthcoming. Our results are not yet conclusive. GA2Ms show some promise for distilling binary classification teachers, but not yet regression. GBTs are not "directly" interpretable but may be promising for regression teachers. GA2M models may provide a computationally viable alternative to additive decomposition methods for global function approximation.

Flying cars have always been a goalpost of the future, but last year companies like Toyota, Airbus, DeLorean and Volvo's parent company invested in or announced plans to get their own units flying soon. If you wanted to get in on the ground floor of tomorrow's transportation, you might try joining the first class of'flying car engineers' in a new nanodegree program at Udacity fronted by Sebastian Thrun, the former leader of Google's self-driving car program. Thrun has quite a pedigree as a founder of Udacity himself along with the Kitty Hawk prototype flying'car,' but the rest of the course's instructors are likewise impressive. They include MIT professor Nicholas Roy, founder of the Alphabet-backed Project Wing whose drones air-delivered burritos to Australians last October; Aerospace professor at University of Toronto Angela Schoellig; And lastly the founder of Kiva Systems (now Amazon Robotics), Raffaello D'Andrea. The course itself aims to educate engineers on both robotics and aerospace concepts to understand particular demands of'flying cars.'

Halo announced today the worldwide release of HaloBoost, Halo's proprietary demand forecasting engine that leverages proven Machine Learning algorithms. HaloBoost combines Machine Learning methods to improve forecast accuracy over time, a high-speed modeling workflow to improve analyst productivity and knowledge discovery, and a simple, scalable method to introduce external factors like pricing, promotion, social media, and weather predictors. "Manufacturers, Distributors, and Retailers have been seeking tools that can provide simplification in the forecasting process to improve accuracy and throughput, and we've responded by introducing our most powerful modeling engine, HaloBoost . Traditional approaches are limited in their ability to maximize forecast accuracy without significant analyst effort across broad and sparse data dimensions such as regions, points-of-sale, and SKU-level granular forecasts. Our proprietary modeling workflow effectively uses the computer to simulate a large team of forecast experts, working in real-time, to find the best result across a broad range of forecast scenarios," said Bill Panak, Ph.D. Vice President of Data Sciences, Halo.

It's that time of year where every summer music festival announces its lineup with a poster filled with band names. It's a tried and true formula, and one that is ripe to be made fun of with a little humor and an artificially intelligent neural network trained on a data set of thousands of band names. Announcing your 2018 COACHELLA LINEUP, generated by a neural network trained on thousands of band names: https://t.co/EskuBWOdfy All these computer-generated names are good, but some of them are more than good. We've separated those out, and recognized them in the following three categories: Digg is what the internet is talking about, right now. It's also the website you are currently on.

Machine Learning is a hot topic these days, as can be seen from search trends. It was the success of Deepmind and AlphaGo in 2016 that really brought machine learning to the attention of the wider community and the world at large. Yet it's a success that followed a long preamble that includes recent advances in three key areas: hardware, particularly GPUs (ideally suited to the vector and matrix based mathematics usually required in machine learning); data, due to the accessibility of larger and larger datasets; and algorithms and techniques, as deep learning research breakthroughs like those described in Krizhevsky, Sutskever and Hinton's landmark paper began to demonstrate best-of-breed results on benchmark challenges. So it's not just hype, and as IT engineers it's worth our while to gain better understanding of it. But the field can seem rather daunting to a newcomer due to all the math, statistics and algorithms involved.