On February 21st 2017, MIT Technology Review—10 Breakthrough Technologies conference was held in Beijing. Jason Pontin, the editor in chief of MIT Technology Review, previewed some of the release of the magazines annual 10 breakthrough technologies issue. Apparently, cognitive technology will be the major theme in the near future. 2016 was huge for advancements in artificial intelligence and machine learning. 2017, however, may well deliver even more.
Initially, the reinforcement theory and behavior analysis was first introduced by comparative psychologist Edward Thorndike in the 20th century. Reinforcement Learning (RL), then, is a type of machine learning, and it was inspired by behaviorist psychology. RL allows machines and software agents to automatically determine the ideal behavior within a specific context, in order to maximize its performance (or cumulative rewards). AlphaGo’s historic victory against one of the best Go players of all time, LeeSedol, was a landmark for the field of AI, and especially for the technique known as deep reinforcement learning. Essentially, RL mimics the way that animals learn how certain behaviors tend to result in a positive or negative outcome. In this sense, machines are able to learn without instruction or even explicit examples.
In 2017, it is very likely that we can anticipate some attempts to apply RL to problems such as automated driving and industrial robotics. In fact, Google has already tested several uses of RL to make its data centers more efficient. Still, the RL approach remains experimental—it requires testing algorithms under different simulated environments, which is time consuming.
On October 2016, an autonomous truck operated by Uber subsidiary Otto delivered 51,744 cans of Budweiser beer from Fort Collins to Colorado Springs. According to American Transportation Research Institute (ATRI), autonomous vehicle technology will penetrate all aspects of the trucking industry, and self-driving trucks will dramatically change its business model. In the U.S., heavy trucks represent about 5% of the total vehicle population, yet consumes 20% of total transportation fuel. The chief executive of Otto, Anthony Levandowski, shows that 6.9 billion hours and 3.1 billion gallons of fuel are wasted every year, thus, the opportunities to save big are tempting. Self-driving trucks have control systems that are capable of analyzing sensory data to distinguish between different cars on the road, which is useful in planning a path to the desired destination.
However, like every other technologies, self-driving trucks encounter several main obstacles such as implementation of legal framework and regulation, as well as the concerns about the resulting loss of driving-related jobs in the road transport industry. Specifically, since truck driver is the largest job in 29 states in the U.S., questions are rising—where does self-driving truck leave the 3.5 million truckers whose livelihoods depend on the need for a human behind the wheel? And what of the millions more whose livelihoods depend on the truckers coming up and down the country, stopping for food, drinks and sleep?
The 360-Degree Selfie
Koen Hufkens is the one who first started the 360-Degree Selfie. He is an ecologist at Harvard University’s Richardson Lab, and he broadcasted a live feed of continuously refreshing, 360-degree still images (one every 15 minutes) from a spot deep in the middle of the Harvard Forest. Hufkens installed a consumer-grade Ricoh Theta S 360-degree camera in the forest. Then, using desktop web browser, people can click and drag around the feed to see different angles of the woods from a height of about 5 feet.
Initially, 360-degree cameras were only utilized on mapping and street views. Over the years, 360-degree cameras have found applications in various domains such as panoramic photos, interactive videos, and filmmaking. And now, 360-degiree cameras are considered to be revolutionary VR products. It can record a video simultaneously from all angles of a scene. 360-degree camera also creates a new format of filming, which has tremendous novelty value. “I discovered that 360-degree video is much more intimate than I expected…I’m not pointing a camera at anyone—I set the camera down and people forget about it, and I’m able to capture these powerful experiences in a very organic way” says Daniel Chase. Daniel, who is a TV producer, is currently filming the world’s first 360-degree virtual reality TV travel documentary—Chasing The World.