Neural Information Processing Systems http://nips.cc/

We propose a multi-step inertial Forward-Backward splitting algorithm for minimizing the sum of two non-necessarily convex functions, one of which is proper lower semi-continuous while the other is differentiable with a Lipschitz continuous gradient. We first prove global convergence of the algorithm with the help of the Kurdyka-Łojasiewicz property. Then, when the non-smooth part is also partly smooth relative to a smooth submanifold, we establish finite identification of the latter and provide sharp local linear convergence analysis. The proposed method is illustrated on several problems arising from statistics and machine learning.

The Insurance Fraud Bureau (IFB) recently teamed up with data analytics firm SAS to launch a new fraud prevention platform, which facilitates information sharing between insurers and authorities to identify and stop fraudulent activity. According to the IFB, the UK's insurance industry loses a staggering £1.3 billion a year due to fraud, with over 100,000 reported instances. However, the true scale of the crime is likely to be much larger with many events going unreported. To learn more about the fraud prevention platform and the industry's prospects in the battle against fraud, Insurance Business spoke with Ben Fletcher (pictured), director of the Insurance Fraud Bureau. "Fraudsters are always becoming more sophisticated and organised in their techniques and move between products, sectors and tactics," he said.