Engineers at Swiss Center for Electronics and Microtechnology have developed a new machine-learning method capable of cutting energy use, as well as allowing artificial intelligence (AI) to complete tasks that were once considered too sensitive. 

Reinforcement Learning Limitations

Reinforcement learning, where a computer continuously improves upon itself by learning from its past experiences, is a major aspect of artificial intelligence. However, this technology is oftentimes difficult to apply to real-life scenarios and situations, such as training climate-control systems. Applications such as this are not able to deal with drastic changes in temperatures, which would be brought on by reinforcement learning. 

This exact issue is what the CSEM engineers set out to address, and that is when they came up with the new approach. The engineers demonstrated that simplified theoretical models could first be used to train computers, and then they would turn to real-life systems. This allows for the machine learning process to be more accurate by the time it reaches the real-life system, learning from its previous trial-and-errors with the theoretical model. This means that there will be no drastic fluctuations for the real-life system, solving the example issue with climate-control technology. 

Pierre-Jean Alet is head of smart energy systems research at CSEM, as well as co-author of the study. 

“It’s like learning the driver’s manual before you start a car,” Alet says. “With this pre-training step, computers build up a knowledge base they can draw on so they aren’t flying blind as they search for the right answer.”

Energy Cuts

One of the most important aspects of this new method is that it can cut energy use by over 20%. The engineers tested the method on a heating, ventilation and air conditioning (HVAC) system, which was located in a 100-room building. 

The engineers relied on three steps, the first of which was training a computer on a “virtual mode.” This model was constructed through simple equations explaining the behavior of the building. Real building data such as temperature, weather conditions and other variables were then fed to the computer, which resulted in more accurate training. The last step was to allow the computer to run the reinforcement learning algorithms, which would eventually result in the best approach forward for the HVAC system. 

The new method developed by the CSEM engineers could have big implications for machine learning. Many applications that were once thought to be “untouchable” by reinforcement learning, like those with large fluctuations, could now be approached in a new manner. This would result in lower energy usage, lower financial costs and many other benefits. 

The research was published in the journal IEEE Transactions on Neural Networks and Learning Systems, titled “A hybrid learning method for system identification and optimal control.” 

The authors include: Baptiste Schubnel, Rafael E. Carrillo, Pierre-Jean Alet and Andreas Hutter. 

The post Engineers Develop New Machine-Learning Method Capable of Cutting Energy Use appeared first on Artificial Intelligence.

Cornell University engineers and researchers are designing the nation’s first statewide Internet of Things public infrastructure.

Thanks to a $1.5 million grant awarded by the National Science Foundation, Cornell faculty will collaborate with community partners around New York – through Cornell Cooperative Extension (CCE) in each county – to set up networks based on low-power wide-area network (LPWAN) technology, a form of low-frequency radio.

“We aim to create a public Internet of Things model that works here and then becomes replicable for other states,” said Max Zhang, engineering professor and the project’s principal investigator (PI). “We want to provide universal network coverage, ensure data privacy, promote responsible data-sharing, scale up successful Internet of Things implementations and spur technology innovation in underserved areas.”

The Internet of Things (IoT) refers to everyday items being connected digitally. For instance, Zhang said, an IoT-enabled home thermostat can be controlled from a smartphone via Wi-Fi connections.

A public LPWAN network targets long-range, low-power and low-bandwidth applications. Examples include utility companies reading meters from a distance, government agencies observing traffic remotely, and farmers using crop or livestock sensors in fields or barns. Towns can develop road and flood monitoring to protect civic infrastructure, providing vital real-time information via networked connections.

Rural areas long have been plagued by poor cellular connections and limited broadband access. Telecommunication companies are reluctant to invest in rural areas, citing high capital costs and few customers.

“Lack of access to networked technology contributes to social, educational, informational and economic disparities,” said co-PI Lee Humphreys, associate professor of communication. “An Internet of Things contributes to greater social and economic opportunities, such as for young students to complete school assignments based on web information or for households to pay bills electronically.”

According to Zhang, developing new networked technologies can leapfrog traditional wired broadband and create innovation.

“You need to create a reliable Internet of Things infrastructure to handle a digital world; otherwise you’re in a cyber desert,” he said. “This is an opportunity for rural communities. You cannot have a digital revolution in digital darkness.”

The post Cornell University Engineers to design first public, statewide ‘Internet of Things’ appeared first on Artificial Intelligence.

Today, with the tremendous growth of data, businesses need an effective team of data scientists to get the real and actionable value of their data. Without the expertise of how to convert cutting-edge technology into significant insights, big data is nothing. Most data scientists have advanced knowledge and training in statistics, math, and computer science. They have vast experience that extends to data visualization, data mining, and information management. However, as businesses strive to integrate new data management and deliver actionable insights, they must ensure their data science and engineering teams work hand in hand.

As data scientists are capable to dig out new ideas and information from the data their company mines regularly, engineers involve in preparing data and develop, tests and maintain complete architecture.

Data Scientists are accountable for analyzing and interpreting intricate digital data, then carry out data analytics and optimization using machine learning and deep learning to deliver valuable insights. On the other hand, engineers are tasked to create data pipelines at scale. This involves incorporating various big data technologies. They are also tasked with determining which tools are right for the job. They have an in-depth understanding of data technologies and frameworks and how to merge them with data pipelines.

Developing a data pipeline is not an easy task as it requires advanced knowledge of production programming frameworks. Though a data scientist can be able to acquire these skills, but it is not the most efficient use of this resource. Data scientists are not engineers who create production systems, build data pipelines, and unveil machine learning outcomes.

Enabling Data Scientists and Engineers to Work Collaboratively

Data scientists and engineers have various distinct routine concerns. However, many businesses make mistakes pertaining to align the skillsets of both with the actual job title. Thus, positioning both roles to extract actionable insights from data and drive true business value, there is a need to get them to comprehend each other’s terminology. The purpose behind this is to enable both teams to speak the same language and build trust through communication.

By providing cross-training to both data scientists and engineers, organizations can fortify shared learning and break down blockades. Through this, data scientists can learn to write prototypes in production languages, while engineers learn the basics of data science where they can understand how the models work.

Businesses can even concentrate on more deliberated aspects like clean, easy-to-deploy code when their data science and engineering teams speak the same language. Data scientists’ iterative and experimental style of workflow can be muddled to an engineer when they are in the early stage of working on a project. So, if code from the experimentation or prototyping phase is conceded on to engineers, companies can easily conquer the roadblock.

Furthermore, in order to boost value from clean code, productization of it internally can create an environment where both data scientists and engineers can lean on their strengths.

The post BRIDGING THE GAP BETWEEN DATA SCIENTISTS AND ENGINEERS appeared first on Artificial Intelligence.