A lot of technologies in the current times are quite complex to use and may appear impractical due to added expenses and other factors. This might be true in case of Artificial Intelligence, wherein several such factors exist as obstacles for the AI systems to be deployed properly. But, this may soon change as according to the latest AI news, a new technology in the form of Deep-Learning AI is developed which may cause AI to work without the use of Internet and Cloud services. This can easily lower the costs of operations and the process complexity, this being feasible enough for long-term use.
This deep-learning AI software could be suitable for utilization in various devices ranging from smartphones used by the common man to robots working in energy-intensive environments. Alexander Wong, a professor at the University of Waterloo in Canada says that the potential for such a technology is immense and can surely improve understanding of AI which is currently present. In this way, data processing and transmission costs can also be lowered along with improving privacy and decreasing overall expenses.
The deep-learning AI software mimics the human brain by processing data through multiple layers of artificial neurons. This process does not require much power, memory and energy to function. However, the new deep-learning AI adapts itself quickly to any changes that take place in the computing process.
According to another researcher Mohammad Javad Shafiee, who also is a professor at the university, the networks evolve themselves through several generations. During this processes, the networks become smaller so that they can survive in diverse environments. The deep-learning AI was prepared by integrating a chip embedded in a device such as a smartphone. This AI could then operate its speech-activated virtual assistant and other intelligent features, greatly reducing the need for operating without internet service, and hence data usage costs.
For more exciting tech and AI related news, keep reading iT Munch.