Ep. 2: Where Deep Learning Goes Next - Bryan Catanzaro, NVIDIA Applied Deep Learning Research by The AI Podcast published on 2016-12-07T04:04:41Z Bryan Catanzaro, vice president for applied deep learning research at NVIDIA, talks about how we know an AI technology is working, the potential for AI-powered speech, and where we’ll see the next deep learning breakthroughs. Genre Technology Comment by Mohit Hooda yes 2019-08-08T20:41:08Z Comment by machinelrn Yes, wrt to humans being better in reasoning and context. ML/DL techniques can be used for both retrieval-based or generative models, but research seems to be moving into the generative direction. FYSA Here's a good chat about ramping up on ML in an hour https://soundcloud.com/machinelrn/become-a-machine-learning-expert-in-under-an-hour 2018-03-09T23:38:42Z Comment by anoop vijayan nice one! keep up the good work ! 2017-06-12T07:49:01Z Comment by TrueFurby There is this saying that: "We have been born too late to explore the earth and too soon to explore the universe".. However I think there is something far more incredible coming very soon.. something none of us can imagine.. something beyond our grasp.. and will come at ever faster rate 2017-05-02T08:06:09Z Comment by TrueFurby very excited about this.. even developing games will be much faster.. and later there will probably be companies owning such neural net / AI which creates entire new games on its own 2017-05-02T07:45:41Z Comment by badradio data collective 2017-01-19T12:58:55Z Comment by Juan Manuel Martín Castillo I really like this format for Pro Content on NVIDIA. Good Job guys! 2016-12-09T10:36:12Z Comment by subjectvobjectv Creating AI that can deal with emotional needs is a touchy, touchy grey area. Get ethicists on board at NVIDIA. 2016-12-07T22:01:35Z Comment by subjectvobjectv Here's another way of putting this: People smart enough to know what the AI is doing will -NOT- gain the emotional benefits. Why? Because the AI is being dishonest. 2016-12-07T21:58:19Z Comment by subjectvobjectv Do you guys have ethicists or philosophers on board for such sensitive issues? These AIs will be making money for people. Conflict of interest. 2016-12-07T21:20:52Z Comment by subjectvobjectv This is a legitimate moral threshold. The question that must be asked here is whether the AI relies on fooling humans or treating humans as emotional objects. 2016-12-07T21:18:02Z Comment by subjectvobjectv Interesting conversations stem from interest in belonging and discovering fears and culturally contextual factors. The only way to create extensive conversations with deep learning is to also give the AI a history and personality. 2016-12-07T21:10:22Z