In the EXPONENTIAL MINDS’ Artificial Intelligence Bulletin – 2030, the End of Personal Cars we look at how 2030 will mean no more personal cars, Apple’s new acquisition, strange loops in deep learning, smart cities,
Death spiral for cars. By 2030, you probably won’t own one
By 2030, you probably won’t own a car, but you may get a free trip with your morning coffee. Transport-As-A-Service will use only electric vehicles and will upend two trillion-dollar industries. It’s the death spiral for cars.
A major new report predicts that by 2030, the overwhelming majority of consumers will no longer own a car – instead they will use on-demand electric autonomous vehicles.
By 2030, within 10 years of regulatory approval of autonomous electric vehicles (A-EVs), the report says, 95 per cent of all US passenger miles traveled will be served by on-demand, autonomous, electric vehicles that will be owned by fleets rather than individuals.
The provision of this service may come virtually free as part of another offering, or a corporate sponsorship. Imagine, for instance, paying a token sum for a ride into town after buying a latte for $4.50. Or getting a free ride because the local government has decided to make transport easier.
Read more at Renew Economy
Apple Just Acquired This Little-Known Artificial Intelligence Startup
Apple has acquired a data mining and machine learning company Lattice.io, according to multiple sources familiar with the matter.
Apple confirmed the acquisition exclusively to Fortune via telephone on Saturday, and provided the following statement via email: “Apple buys smaller technology companies from time to time, and we generally do not discuss our purpose or plans.”
There isn’t much information publicly available about Lattice, but according to the company’s CrunchBase profile, the startup was born out a Stanford research project called DeepDive. The company’s technology appears to use machine learning to parse through databases or the web to answer queries.
Lattice was co-founded by Chris Re, a professor of computer science at Stanford, and Michael Cafarella, a professor of computer science at the University of Michigan. Cafarella was the co-creator of Hadoop, a widely used big data processing technology. Cafarella was also previously an engineer at telecommunications company TellMe Networks, which was bought by Microsoft in 2007 for $800 million.
According to this 2015 profile on Re, the professor’s Deep Dive program is able to understand “dark data,” which provides information within images or illustrations.
Read more at FORTUNE
The Strange Loop in Deep Learning
Where he describes this self-referential mechanism as what describes the unique property of minds. The strange loop is a cyclic system that traverses several layers in a hierarchy. By moving through this cycle one finds oneself where one originally started.
Coincidentally enough, this ‘strange loop’ is in fact is the fundamental reason for what Yann LeCun describes as “the coolest idea in machine learning in the last twenty years.”
Loops are not typical in Deep Learning systems. These systems have conventionally been composed of acyclic graphs of computation layers. However, as we are all now beginning to discover, the employment of ‘feedback loops’ are creating one of the most mind-boggling new capabilities for automation. This is not hyperbole, this is happening today where researchers are training ‘narrow’ intelligence systems to create very capable specialist automation that surpass human capabilities.
Read more at Medium
When Artificial Intelligence Rules the City
A report by a panel of leading experts on technology, business, and cities takes a deep dive into the changes that will come about as a result of one key new technology—artificial intelligence.
The panel was chaired by Peter Stone of University of Texas at Austin along with researchers from Rethink Robotics, Allen Institute for AI, Microsoft, and academics from Harvard, MIT, Johns Hopkins, Columbia, UC Berkeley, and other universities from around the world. Their study, Artificial Intelligence and Life in 2030, outlines the dramatic impact artificial intelligence (AI) is having and will continue to have for our cities and the way we live and work in them over the next couple of decades. It outlines the implications of several key dimensions of AI, including:
- Large-scale learning or algorithms that crunch ever-larger datasets
- Deep learning procedures that recognize images, video, audio, speech, and language
- Reinforcement learning that shifts from pattern recognition to experience-driven decision-making
- Robotic devices that can physically interact with environments and people
- Computer vision that allows computers to see and perform tasks better than people
- Natural language processing that does more than react to requests—it communicates through speech
- Collaborative systems, crowdsourcing, and human computation
- Algorithms and computational tools that can apply economic and social data to realign incentives for people and businesses
- The “Internet of Things” that networks appliances, vehicles, buildings, and cameras
- Neuromorphic computing that mimics biological neural networks to improve the efficiency and robustness of computer systems
The report outlines what these technologies mean for cities and raises deep policy (and downright philosophical) questions about their impact across several areas of urban life.
Read more at CityLab
A famous venture capitalist predicts big banks will fall first to artificial intelligence
Wall Street will be one of the first and largest industries to be automated by artificial intelligence, predicts Kai-Fu Lee, China’s most famous venture capitalist and former Microsoft and Google executive. Lenders, money managers, and analysts—any jobs that involve crunching numbers to estimate a return—are at risk.
“Banks have the curse of the baggage they have, like Kodak letting go of film,” Lee says. “Their DNA is all wrong.”
Lee’s VC firm, Sinovation Ventures, has started to invest in this space by financing Smart Finance Group, a company which algorithmically determines eligibility for payday loans. Lee expects the company’s algorithms to pay out 30 million loans this year, giving the company scale that would never have been achievable when hiring humans to do the same job. That core technology would be easily applicable to other kinds of loans and financial decisions.
As for the big banks that dominate now, the venture capitalist predicts they will be outmaneuvered by smaller startups able to deploy new technology much faster.
Read more at Quartz