Computer Vision News - January 2022

25 AI Spotlight News Machines That See theWorldMore Like Humans D o Computer vision systems sometimes make inferences about a scene that fly in the face of common sense. This is particularly problematic in systems requiring the highest accuracy, like autonomous driving . Toovercome these errors, MIT researchers have developed a framework using probabilistic programming that helps machines see the world more like humansdo. TheirnewAI learns toperceive real-world objects from just a few images, and perceives scenes in terms of these learned objects, including apparently new concepts like “uncertainty ” and rules that don’t hold. Read More PyTorch vs TensorFlow in 202 2 Should you use PyTorch vs TensorFlow in 2022? Spoiler alert: answers ahead! As you probably expect, the PyTorch vs TensorFlow debate does not have a single correct answer, it all depends on what you want to do . If you perform Deep Learning engineering in an industry setting, you’re likely using TensorFlow and should probably stick with it. If you’re a researcher , you’re almost certainly using PyTorch, and for now you should likely stick with it: PyTorch is the de facto research framework. If you’re looking to make a career change , either PyTorch or TensorFlow is a good option. Read More AI GoneRogue: 6TimesAIWent TooFar We all know a few AI horror stories . Some of them were disclosed here on Computer Vision News. Here is a short list, it’s a nice reading. There might be a couple that you didn’t know about, like when Sophia was asked whether she wanted to destroy humans and replied: “OK. I will destroy humans!” . Or like when two virtual assistant robots quickly turned to throwing insults and threats at each other. Some stories you already know, but it’s always interesting to know how they ended. The article concludes by asking: Rogue AI - What’s Our Best Defense? Don’t expect toomuch of an answer. ReadMore