You could say that the Magnifi 2 is the soundbar-iest soundbar. The 2.1-channel system pulls out all the stops to pull off a multi-speaker experience from its compact design, fully embodying the soundbar creed. To that end, it does as well as could be expected with the tools Polk has provided, and the combination of sheer brute force and clear detail the bar musters from its limited armory is impressive. Nearly every rival at this price adds at least a center channel for dialogue. What's more, the Magnifi 2 can't decode high-end audio formats like Dolby TrueHD or DTS-HD Master Audio (let alone popular 3D formats like Dolby Atmos), leaving it to dwell somewhere in limbo between entry-level bars like Vizio's V21 and multi-channel Atmos powerhouses like the Sonos Arc.

Uber will not be held criminally liable in the fatal crash last year in Tempe, Arizona, in which a self-driving vehicle struck and killed a pedestrian, a county prosecutor announced Tuesday. Yavapai County Attorney Sheila Polk said Uber won't face criminal charges in the March 2018 crash -- believed to be the first fatality in the U.S. involving a self-driving vehicle. Polk said her office concluded that video of the crash likely didn't accurately depict the collision and recommended that Tempe police seek more evidence. DRIVER IN FATAL SELF-DRIVING UBER CRASH WAS WATCHING'THE VOICE,' INVESTIGATORS SAY It's not known whether prosecutors are considering charges against the driver. Dashcam video released by the Tempe Police Department last year showed an interior view of Uber backup driver Rafael Vasquez in the moments before the crash.

PHOENIX – Prosecutors announced Tuesday that they didn't find evidence to criminally charge Uber in the crash that killed a woman a year ago in Tempe. But it is leaving possible criminal charges against the autonomous car's operator back in Maricopa County officials' hands. Yavapai County Attorney Sheila Sullivan Polk's Office took the case at the request of Maricopa County Attorney Bill Montgomery after he cited a potential conflict of interest. Polk in a Monday letter to Montgomery said her office would recommend that Tempe police further investigate to help Montgomery's office determine if any other charges should be filed against the driver. A report the Tempe Police Department released in June revealed 44-year-old Rafaela Vasquez, the operator of Uber's self-driving vehicle, was watching "The Voice" via a streaming service when the autonomous car hit 49-year-old Elena Herzberg on March 18 as she crossed a street outside of a crosswalk with her bike.

To answer the question posed in the subhed, no, Polk didn't create the Command Bar by taking a hole saw to the middle of its chassis and dropping an Amazon Echo Dot in there. But the feature is unquestionably designed to look like that, and this soundbar does support Amazon's Alexa digital assistant. Soundbars are hot these days, and they run the gamut from sub-$100 cheapies to the ludicrously priced, but wow-it's-impressive Creative X-Fi Sonic Carrier. Polk is going after the better-than-entry-level market with this $300 speaker. That makes it $100 cheaper than the recently reviewed Sonos Beam, and it comes with a nicely matched subwoofer.

Doug Polk is a professional poker player who's won millions of dollars, mostly at heads-up Texas Hold Em No Limit. He was part of a team that recently beat an artificial intelligence bot programmed by MIT students. When they're not at the tables collecting cash, Polk and fellow poker pro Ryan Fee run Upswing Poker, a site that offers training to everyday players who want to improve. Since so much Wall Street trading is conducted by computer programs these days, I figured he might have some wisdom about going up against AI for the few remaining humans who discretionary trade. John Navin: Like many human, discretionary traders on Wall Street, you've gone up against an artificial intelligence bot.

Despite their attractiveness, popular perception is that techniques for nonparametric function approximation do not scale to streaming data due to an intractable growth in the amount of storage they require. To solve this problem in a memory-affordable way, we propose an online technique based on functional stochastic gradient descent in tandem with supervised sparsification based on greedy function subspace projections. The method, called parsimonious online learning with kernels (POLK), provides a controllable tradeoff? between its solution accuracy and the amount of memory it requires. We derive conditions under which the generated function sequence converges almost surely to the optimal function, and we establish that the memory requirement remains finite. We evaluate POLK for kernel multi-class logistic regression and kernel hinge-loss classification on three canonical data sets: a synthetic Gaussian mixture model, the MNIST hand-written digits, and the Brodatz texture database. On all three tasks, we observe a favorable tradeoff of objective function evaluation, classification performance, and complexity of the nonparametric regressor extracted the proposed method.