The state-space models (SSMs) are widely utilized in the analysis of time-series data. SSMs rely on an explicit definition of the state and observation processes. Characterizing these processes is not always easy and becomes a modeling challenge when the dimension of observed data grows or the observed data distribution deviates from the normal distribution. Here, we propose a new formulation of SSM for high-dimensional observation processes. We call this solution the deep direct discriminative decoder (D4). The D4 brings deep neural networks' expressiveness and scalability to the SSM formulation letting us build a novel solution that efficiently estimates the underlying state processes through high-dimensional observation signal. We demonstrate the D4 solutions in simulated and real data such as Lorenz attractors, Langevin dynamics, random walk dynamics, and rat hippocampus spiking neural data and show that the D4 performs better than traditional SSMs and RNNs. The D4 can be applied to a broader class of time-series data where the connection between high-dimensional observation and the underlying latent process is hard to characterize.

Advertising, particularly online advertising, isn't a surefire way to bolster business. A report from ecommerce analytics platform Glew drives the point home: In 2015, 75% of retailers that spent at least $5,000 on Facebook ads ended up losing money on those ads, with the average return on investment landing around -66.7%. A 2018 survey of marketers by Rakuten Marketing found that companies waste an estimated 26% of their budgets on inefficient ad channels and strategies. Jaleh Rezaei, the CEO of Mutiny, believes that the problem doesn't lie with the ads themselves. Rather, she pegs it on static, templated websites that don't match the personalization delivered by ads.

NEW YORK – From the vast deserts of Saudi Arabia to the crowded neighborhoods of Beirut, a drone war has taken flight across the wider Middle East, raising the stakes in the ongoing tensions between the U.S. and Iran. In the year since President Donald Trump withdrew America from Iran's nuclear deal, there's been an increasing tempo of attacks and alleged threats from unmanned aircraft flown by Tehran's and Washington's allies in the region. The appeal of the aircraft -- they risk no pilots and can be small enough to evade air-defense systems -- fueled their rapid use amid the maximum pressure campaigns of Iran and the U.S. As these strikes become more frequent, the risk of unwanted escalation becomes greater. The U.S. military nearly launched airstrikes against Iran after a U.S. military surveillance drone was shot down in June. Meanwhile, Israeli fighter jets attack targets in Syria on an almost weekly basis, including on Saturday night.