The recently launched LinkedIn Salary product has been designed with the goal of providing compensation insights to the world's professionals and thereby helping them optimize their earning potential. We describe the overall design and architecture of the statistical modeling system underlying this product. We focus on the unique data mining challenges while designing and implementing the system, and describe the modeling components such as Bayesian hierarchical smoothing that help to compute and present robust compensation insights to users. We report on extensive evaluation with nearly one year of de-identified compensation data collected from over one million LinkedIn users, thereby demonstrating the efficacy of the statistical models. We also highlight the lessons learned through the deployment of our system at LinkedIn.

You've probably seen the Watson commercials, where what looks like a sentient box interacts with celebrities like Bob Dylan, Carrie Fisher, and Serena Williams; or doctors; or a young cancer survivor. Maybe you caught the IBM artificial intelligence technology's appearance in H&R Block's Super Bowl commercial starring Jon Hamm. "It is one of the most powerful tools our species has created. It helps doctors fight disease," Hamm says. "It can predict global weather patterns. It improves education for children everywhere. And now we unleash it on your taxes."

Appier, a Taiwanese startup that helps companies harness artificial intelligence to make marketing decisions, announced today that it has raised a $33 million Series C round from an impressive roster of Asian investors. They are SoftBank Group Corp., Line Corp., Naver Corp., EDBI (the Singapore Economic Development Board's corporate investment arm) and Hong Kong-based financial services firm AMTD Group. This brings Appier's total funding so far to $82 million. Lead investors in its previous rounds included Sequoia Capital and Pavilion Capital. Appier co-founder and CEO Chih-Han Yu says its Series C will be used to grow its engineering and research and development teams in countries outside of Taiwan, including Singapore.

This question of who a vehicle should kill when placed in a situation where every outcome would end in death is often called the "Trolley problem." It's an ethical debate that's lasted more than 60 years: Ethicists riddle each other with questions of whether it's excusable to kill two elderly people to save one child, or save a pregnant woman while killing a man and a child, and on and on. MIT even made a game to test your own ethical predisposition in the situation.

There is no denying the fact that Artificial Intelligence is the breakthrough technology of recent times. The machines have come a long way from assisting humans in mechanical operations to performing smarter tasks using cognitive intelligence. Every day, we are coming across interesting applications of AI. The ability of Deep Learning algorithms to learn and predict efficiently has opened the doors of possibilities. Nowadays, AI is impacting many other areas as well.

The rise of automation and artificial intelligence has prompted all kinds of questions about what the future holds for human workers. However, a new study conducted by the World Economic Forum suggests that millennials aren't too concerned about technology's impact on the world of work -- in fact, they largely hold the opinion that recent advances will lead to more jobs, not less. A total of 24,766 participants were grilled on topics ranging from the global economic outlook to the role of technology in society to produce the 2017 edition of the Global Shapers Survey. It's worth noting that participants who were identified as part of the low human development group on the Human Development Index were less confident in technology's capacity to create jobs, rather than take them away. Only 66.7 percent voted for creation, the lowest proportion of any group on the scale. This report certainly seems to suggest that by and large, younger people don't fear the rise of automation.

Raphael Goldsworthy is the Managing Director of Better Boards Australasia, convenor of the largest annual gathering of Not-For-Profit (NFP) and For-Purpose directors in the Southern Hemisphere, the Better Boards Conference. Raphael has spent almost 10 years working closely with, and curating educational programs for, NFP directors, boards and executives. Raphael has a deep interest in the intersection of technology, decision making, investment and behavioural economics. He regularly writes and speaks on technology in the boardroom, decision making and biases, governance, leadership and related NFP matters.

March 6, 2017 Written by: Susan C. Daffron If you've had to deal with managing streaming data, maybe you've heard of PubNub. Now it's easy to add Watson-powered machine intelligence to those streams with BLOCKS, a feature of the PubNub Data Stream Network (DSN) that makes the network programmable. Using BLOCKS, developers can easily deploy functions on the PubNub network to modify messages without the need to manage their own infrastructure. In a new episode of the Building with Watson webinar series, Josh Marinacci, Head of Developer Relations at PubNub demonstrates how he used the Watson Conversation PubNub BLOCK to build a geology-themed chatbot called Mr. Rockbot. When you're building a chatbot, you need to remember that a chatbot involves constant communication between the user and your bot. To tie these elements together, you'll need a real-time, low-latency and high security infrastructure.

Neural Information Processing Systems (NIPS) is a top-tier annual conference in machine learning. The 2016 edition of the conference comprised more than 2,400 paper submissions, 3,000 reviewers, and 8,000 attendees, representing a growth of nearly 40% in terms of submissions, 96% in terms of reviewers, and over 100% in terms of attendees as compared to the previous year. In this report, we analyze several aspects of the data collected during the review process, including an experiment investigating the efficacy of collecting ordinal rankings from reviewers (vs. usual scores aka cardinal rankings). Our goal is to check the soundness of the review process we implemented and, in going so, provide insights that may be useful in the design of the review process of subsequent conferences. We introduce a number of metrics that could be used for monitoring improvements when new ideas are introduced.

We consider the problem of noisy matrix completion, in which the goal is to reconstruct a structured matrix whose entries are partially observed in noise. Standard approaches to this underdetermined inverse problem are based on assuming that the underlying matrix has low rank, or is well-approximated by a low rank matrix. In this paper, we propose a richer model based on what we term the "permutation-rank" of a matrix. We first describe how the classical non-negative rank model enforces restrictions that may be undesirable in practice, and how and these restrictions can be avoided by using the richer permutation-rank model. Second, we establish the minimax rates of estimation under the new permutation-based model, and prove that surprisingly, the minimax rates are equivalent up to logarithmic factors to those for estimation under the typical low rank model. Third, we analyze a computationally efficient singular-value-thresholding algorithm, known to be optimal for the low-rank setting, and show that it also simultaneously yields a consistent estimator for the low-permutation rank setting. Finally, we present various structural results characterizing the uniqueness of the permutation-rank decomposition, and characterizing convex approximations of the permutation-rank polytope.