What is Machine Learning that the market is adopting? Machine Learning can be defined in various ways depending on one's understanding, the author or articles referred to. In my definition, machine learning is a branch of artificial intelligence. It provides the systems with the capability to grasp information; learn and act like humans. Machine Learning helps the system improve their learning over time through interactions, observations without being solely programmed.

So you've decided data science is the field for you. More and more businesses are becoming data driven, the world is increasingly becoming more connected and looks like every business will need a data science practice. So, the demand for data scientists is huge. Even better, everyone acknowledges the shortfall of talent in the industry. However, becoming a data scientist does not come easy. It needs a mix of problem solving, structured thinking, coding and various technical skills among others to be truly successful. If you are from a non-technical and non-mathematical background, there's a good chance a lot of your learning happens through books and video courses. Most of these resources don't teach you what the industry is looking for in a data scientist.

Forty percent of HR leaders believe artificial intelligence will help fill the skills gap. That's according to a new study by Learning House and Future Workplace, which surveyed 600 U.S. HR leaders. More than half of those surveyed acknowledged the skills gap and more than a third believe it's harder to fill open positions now than it was in 2017, but some critics say companies are not doing much to fix the problem. The study found that 74 percent of companies are only investing $500 per employee on learning and development. Jeremy Walsh, senior vice president of enterprise learning solutions at Learning House, said he was shocked by the low amount of money being spent on L&D.

Zachary C. Lipton and Jacob Steinhardt have written an outstanding paper, entitled Troubling Trends in Machine Learning Scholarship, that focuses on patterns that are a trend in the scientific literature from the machine learning community: the failure to distinguish between explanation and speculation, the failure to identify the sources of empirical gains, the mathiness (the use of mathematics that obfuscates or impresses rather than clarifies), and the misuse of language. Those are the same issues that we find, PERMANENTLY!, in the failed efforts to define what (machine) intelligence is and is not, since the very origins of AI. Lipton and Steinhardt elaborate on the following possible causal factors: a complacency in the face of progress, the rapid expansion of the community, the consequent thinness of the reviewer pool, and misaligned incentives of scholarship vs. short-term measures of success. They don't stop there, they even provide suggestions for authors, publishers, and reviewers, and conclude that It is tremendously important that each and every AI researcher, practitioner, author, reviewer, publisher, investor, journalist, user, leader, student, educator, and enthusiast is aware of these implications. We will never advance AI in the right direction by carrying flawed scholarship all the way with us.

Brandman University is taking a new approach to adult education, focusing on student competencies and work experience rather than transcripts when deciding which students to admit and when they graduate. Brandman already is working with companies, including Walmart and Discover, to offer employee-education programs. The Irvine, California-based nonprofit university accepts subject matter expertise and experience as course credit, making it easier for working adults to earn college degrees and advance their careers. At most conventional colleges, students must fulfill prerequisite courses to earn admission and a set of required courses to earn a degree. Under the Brandman approach, if an applicant has, say, a 20-year career in finance but no formal coursework in finance, "she can now test out of many course requirements, simply by proving her mastery through standard assessments, writing samples, even work projects," says the university's chief financial officer.

Artificial intelligence is already reshaping society as we know it in both business and consumer realms. Early use cases with Alexa, autonomous vehicles and AI-driven supply chains provide just a glimpse of the disruption that AI is poised to deliver in the near future and for years to come. Yet despite all the AI hype and initial successes, it remains in its infancy. That makes now the ideal time for young people to build the knowledge, skill sets and connections they need to capitalize on the fast-growing market for AI jobs and build a strong AI career. One reason is simply practical. Gartner predicts that AI may eliminate 1.8 million jobs by 2020, yet is on track to create 2.3 million new positions.

How do you teach computers to understand language -- not just transcribe human speech, but actually comprehend what someone is saying? It's one of the grand challenges of AI, and we still don't really know the best way to tackle the problem. Facebook's AI research lab, FAIR, has one idea: teach AIs to understand language by getting them to guide virtual tourists around New York City. FAIR is releasing what it calls Talk the Walk, a dataset designed to be used by other researchers. It's comprised of three elements: small maps of New York City neighborhoods (each a couple of blocks wide), 360-degree photos of the same locations, and sample dialogues of humans guiding one another around these neighborhoods. Basically, it's everything you might need to teach an AI to tackle this task itself.

Games generalize the optimization paradigm by introducing different objective functions for different optimizing agents, known as players. Generative Adversarial Networks (GANs) are arguably the most popular game formulation in recent machine learning literature. GANs achieve great results on generating realistic natural images, however they are known for being difficult to train. Training them involves finding a Nash equilibrium, typically performed using gradient descent on the two players' objectives. Game dynamics can induce rotations that slow down convergence to a Nash equilibrium, or prevent it altogether. We provide a theoretical analysis of the game dynamics. Our analysis, supported by experiments, shows that gradient descent with a negative momentum term can improve the convergence properties of some GANs.

We study a general online linear optimization problem(OLO). At each round, a subset of objects from a fixed universe of $n$ objects is chosen, and a linear cost associated with the chosen subset is incurred. We use \textit{regret} as a measure of performance of our algorithms. Regret is the difference between the total cost incurred over all iterations and the cost of the best fixed subset in hindsight. We consider \textit{Full Information}, \textit{Semi-Bandit} and \textit{Bandit} feedback for this problem. Using characteristic vectors of the subsets, this problem reduces to OLO on the $\{0,1\}^n$ hypercube. The Exp2 algorithm and its bandit variants are commonly used strategies for this problem. It was previously unknown if it is possible to run Exp2 on the hypercube in polynomial time. In this paper, we present a polynomial time algorithm called \textit{PolyExp} for OLO on the hypercube. We show that our algorithm is equivalent to both Exp2 on $\{0,1\}^n$ as well as Online Mirror Descent(OMD) with Entropic regularization on $[0,1]^n$ and Bernoulli Sampling. Under $L_\infty$ adversarial losses, in the Full Information case and Semi-Bandit case, analyzing Exp2 directly, gives an expected regret bound of $O(n^{3/2}\sqrt{T})$, whereas PolyExp yields a regret of $O(n\sqrt{T})$. In the Bandit case, analyzing Exp2 directly, gives an expected regret bound of $O(n^{2}\sqrt{T})$, whereas PolyExp yields a regret of $O(n^{3/2}\sqrt{T})$. This implies an improvement on Exp2's regret bound for these settings because of the equivalence. Moreover, PolyExp is minimax optimal in all the three settings as its regret bounds match the $L_\infty$ lowerbounds in \cite{audibert2011minimax}. Finally, we show how to use PolyExp on the $\{-1,+1\}^n$ hypercube, solving an open problem in \cite{bubeck2012towards}.

Artificial intelligence technology in education is no more the future, it is the present. We are already witnessing the impact of technology in the education sector. Increased uses of digital devices, adaptive learning platforms, and engaging lessons have transformed the learning as well as teaching processes. With the help of artificial intelligence technology, the educators today are working on creating activities that can not only ignite the curiosities among the students, but can make their learning experience a memorable one. Although AI can never replace the human teachers, it has successfully given a whole new meaning to the roles and responsibilities of a teacher.