Nav: Home

Autonomous machines edge towards greater independence

June 08, 2017

Cars that can drive autonomously have recently made headlines. In the near future, machines that can learn autonomously will become increasingly present in our lives. The secret to efficient learning for these machines is to define an iterative process to map out the evolution of how key aspects of these systems change over time. In a study published in EPJ B, Agustín Bilen and Pablo Kaluza from Universidad Nacional de Cuyo, Mendoza, Argentina show that these smart systems can evolve autonomously to perform a specific and well-defined task over time. Applications range from nanotechnology to biological systems, such as biological signal transduction networks, genetic regulatory networks with adaptive responses, or genetic networks in which the expression level of certain genes in a network oscillates from one state to another.

These autonomous systems do not need an external tutor. Nor do they report to a central unit designed to modify what the system must learn depending on their performance. To increase their autonomy, the authors have built in delayed dynamics and a feedback loop with the system's performance. The delayed dynamics provide information on the history of the system, thus presenting the past relationships between its structure and performance. In turn, the feedback loop offers information on the system's actual performance in terms of how close it is to the desired task.

The researchers first successfully applied their approach to a neural network responsible for classifying several patterns, which yielded 66% robustness. These insights can be applied e.g. in analogue electronics, where a piece of hardware can learn a task autonomously without a central processing unit or external control. The team also tested their solution on a system of phase oscillators, which are interesting because the populations of such oscillators display some remarkable synchronisation traits. In such cases, autonomous learning helps to avoid the inherent fluctuations of error normally found in such systems.
-end-
Reference:

Agustín M. Bilen and Pablo Kaluza (2017), Autonomous learning by simple dynamical systems with a discrete-time formulation, Eur. Phys. J. B 90:94, DOI 10.1140/epjb/e2017-70714-7

Springer

Related Autonomy Articles:

Scientists help soldiers figure out what robots know
An Army-led research team developed new algorithms and filled in knowledge gaps about how robots contribute to teams and what robots know about their environment and teammates.
Tension around autonomy increases family conflict at end of life
Conflict within families can be stressful and confusing, and it can lead to feelings of sadness.
Children should help choose the charities their schools and families support
Children as young as four should be given more autonomy to help choose the charities their schools and families support, according to new research from the University of Kent and Canterbury Christ Church University.
Parental support linked to how well millennials transition to college life
Researchers show that how well parents or guardians support millennials' psychological needs prior to their transition to college is an important predictor of their psychological well-being as they adapt to college life.
Work-family conflict hits home
Researchers have long known that sick children can affect a company's bottom line, as employees are distracted or have to take time off to care for their children.
Tenacious and flexible goal pursuit gets older people on the move
Tenacious goal pursuit and flexible goal adjustment have been shown to help maintain psychological well-being despite age related challenges and losses.
Pumping up fitness app features may add muscle to workout commitment
Fitness apps are easy to download and can help motivate people to start workout routines, but that may not be enough to sustain those routines in the long run.
Out of office: New Baylor study examines relationship between stress and remote work
Many US employees believe working from home -- or at least away from the office -- can bring freedom and stress-free job satisfaction.
Disaster relief: How can AI improve humanitarian assistance?
The unique topic of artificial intelligence (AI) for humanitarian assistance and disaster relief (HA/DR) was in the spotlight last week, as leading minds from academia, industry and the federal government met to discuss how modern technology can help victims of disasters around the globe.
Does increased supervision of resident physicians reduce medical errors?
Increased supervision of residents by attending physicians who joined patient rounds didn't significantly reduce medical errors but residents reported decreased autonomy.
More Autonomy News and Autonomy Current Events

Top Science Podcasts

We have hand picked the top science podcasts of 2019.