The Robot Computing Era — Tue links
The below tweet from @estherschindler links to some recent predictions by sci-fi author @David_Gerrold, where he outlines his vision for the future of computing.
Since y'all asked for an update....
— Esther Schindler (@estherschindler) April 3, 2018
SF author @DavidGerrold nailed smartphones in 1999. Here’s his next prediction: “My friend, the robot” https://t.co/I2sfXIizuA
Gerrold knows his stuff — he made a prescient forecast of smartphones back in 1999. He called the devices “PITA” (Personal Information Telecommunciations Agent)…
“The acronym also can stand for Pain In The Ass, which it is equally likely to be, because having all that connectivity is going to destroy what’s left of everyone’s privacy.”
Sounds pretty spot on, no?
In his new essay, Gerrold describes how the next wave of computing will be driven by robots. He suggests that robots will handle many of our day-to-day tasks, surpassing what even human assistants can do to become a kind of “life manager”. (hmm, this sounds like the world of the movie “Her”)
From the essay…
“The more sophisticated a robot’s information processing ability, the more it will develop a personality tuned to the user. It will become a companion. It will become an electronic friend.”
(yup, definitely sounds like “Her”)
“And then there’s this: Robots as romantic partners.”
(ok, now we’re definitely talking “Her”!)
Gerrold speculates that the development of these “true robots” will take at least another decade, but this will probably be a good thing, giving us humans more time to figure out how they should fit into our society.
But as long as these robots help to enhance our humanity rather than provide a substitute, I think we’ll be fine. For example, thinking of computers as bicycles for the mind should serve us well.
And now on to the links…
1. From NVIDIA, here’s DeepRecommender: Deep AutoEncoders for Collaborative Filtering.
(hat tip to the incomparable @jeremyphoward)
2. Interesting thought about smarter people having more neurons than average, but also fewer neural connections — is this brain “regularization”?
Is this a form of regularization (in the machine learning sense) in the brain? -> https://t.co/lJ5TKxJnYm
— John Alberg (@JohnAlberg) May 22, 2018
(key_awaytaking: over-fitting is for under-achievers)
3. Here’s a nice overview of using deep learning for object detection by @ZoeyC17.
(you can also see an example of using MobileNets for object detection in this notebook)
4. @BenedictEvans speculates on whether we’ll ever have an L5 autonomous car.
(farewell_our_robot_overlords: maybe the news won’t become all “machine bites man” headlines after all?)
5. Cool look at a deserted Great Wall of China by photographer Andrés Gallardo Albajar.
(colossal_understatement: follow @Colossal)