Google's DeepMind was reported to help advance Robotics and will pave the way to comprehend how humans learn. This Alphabet subsidiary is geared towards the important progress in artificial intelligence. It is about to release a 3-D virtual world that can benefit students on their experiments.
Shane Legg, co-founder and chief scientist of DeepMind said that they are developing these AI agents that can function on a variety of tasks - observing what happens and looking at the environment. DeepMind Lab, its platform is like a compact 3-D first-person shooter computer game. The artificial intelligence becomes like a floating orb that can identify its surroundings, act and move around. These agents can be molded to do different tasks machine-like that involve receiving positive rewards.
If DeepMind is tailored to fit one's preference it can provide more intricate, complex and visually rich challenges for agents. It, however, entails a bigger and wider range of possible tasks as per Technology Review. Legg further states that "Having AI agents work inside a 3-D environment could also have benefits for developing algorithms to control systems that work in the real world such as industrial robots."
Other experts in the field welcomed DeepMind Lab. According to Ilya Sutskevar, research director and co-founder at OpenAI. Theirs is doing basic research and releasing them in public. The experts said that "The more environments reinforcement learning agents have access to, the faster the field will move forward."
The more platforms for reinforcement learning lead to progress by allowing researchers to try the ideas of others. However, Zoubin Gahrahmani, University of Cambridge professor in the U.K. said that these approaches to reinforcement cannot measure up to the innate human abilities. This is one of the identified downsides.
Goodge's DeepMind is a prodigy's work and is worth trying. Stay tuned for more technological breakthroughs and business news.