Neural learning occurs at dendrite roots, not in synapses. The newly suggested learning scenario indicates that learning occurs in a few dendrites that are in much closer proximity to the neuron, as opposed to the previous notion. … The new learning scenario occurs in different sites of the brain and therefore calls for a reevaluation of current treatments for disordered brain functionality. … In addition, the learning mechanism is at the basis of recent advanced machine learning and deep learning…
A Guardian article last October brings the darker aspects of the attention economy, particularly the techniques and tools of neural hijacking, into sharp focus. The piece summarizes some interaction design principles and trends that signal a fundamental shift in means, deployment, and startling effectiveness of mass persuasion. The mechanisms reliably and efficiently leverage neural reward (dopamine) circuits to seize, hold, and direct attention toward whatever end the designer and content providers choose. The organizer of a $1,700 per person event…
AI system can isolate individuals’ voices from other environmental noise, including other voices. Such a system has many potential uses, both benign and nefarious. The ability is rapidly improving to untangle signals from noise and identify which signals are from which sources. The approach should be able to apply to other kinds of signals too, not only sounds. https://www.newscientist.com/article/2151268-an-ai-has-learned-how-to-pick-a-single-voice-out-of-a-crowd/
An MIT Technology Review article introduces the man responsible for the 30-year-old deep learning approach, explains what deep machine learning is, and questions whether deep learning may be the last significant innovation in the AI field. The article also touches on a potential way forward for developing AIs with qualities more analogous to the human brain’s functioning.
Should it surprise us that human biases find their way into human-designed AI algorithms trained using data sets of human artifacts? Machine-learning software trained on the datasets didn’t just mirror those biases, it amplified them. If a photo set generally associated women with cooking, software trained by studying those photos and their labels created an even stronger association. https://www.wired.com/story/machines-taught-by-photos-learn-a-sexist-view-of-women?mbid=nl_82117_p2&CNDID=24258719
This NY Times article is worth your time, if you are interested in AI–especially if you are still under the impression AI has ossified or lost its way.
TED talk of possible interest: http://www.ted.com/talks/zeynep_tufekci_we_can_t_control_what_our_intelligent_machines_are_learning?utm_source=newsletter_weekly_2016-10-22&utm_campaign=newsletter_weekly&utm_medium=email&utm_content=talk_of_the_week_swipe Comment I posted there: Here is an interdisciplinary “moon-shot” suggestion that we should at least start talking about, now, before it is too late. Let’s massively collaborate to develop a very mission-specific AI system to help us figure out, using emerging genetic editing technologies (e.g., CRISPR, etc.), ideally how to tweak (most likely) species-typical genes currently constraining our capacities for prosociality, biophilia, and compassion, so that we can intentionally evolve into a sustainable species….
“Once the trained CNN [convolutional neural network] showed solid performance in the simulator, we loaded it onto DRIVE PX [vehicle control computer] and took it out for a road test in the car. The vehicle drove along paved and unpaved roads with and without lane markings, and handled a wide range of weather conditions. As more training data was gathered, performance continually improved. The car even flawlessly cruised the Garden State Parkway.”