The idea that “time-traveling” pathogens trapped in ice or hidden in remote laboratory facilities could break free to cause catastrophic outbreaks has inspired generations of novelists and screenwriters. While melting glaciers and permafrost are giving many types of dormant microbes the opportunity to re-emerge, the potential threats to human health and the environment posed by these microbes have been difficult to estimate.

In a new study, Strona’s team quantified the ecological risks posed by these microbes using computer simulations. The researchers performed artificial evolution experiments where digital virus-like pathogens from the past invade communities of bacteria-like hosts. They compared the effects of invading pathogens on the diversity of host bacteria to diversity in control communities where no invasion occurred.

The team found that in their simulations, the ancient invading pathogens could often survive and evolve in the modern community, and about 3 percent became dominant. While most of the dominant invaders had little effect on the composition of the larger community, about 1 percent of the invaders yielded unpredictable results. Some caused up to one third of the host species to die out, while others increased diversity by up to 12 percent compared to the control simulations.

The risks posed by this 1 percent of released pathogens may seem small, but given the sheer number of ancient microbes regularly released into modern communities, outbreak events still represent a substantial hazard. The new findings suggest that the risks posed by time-traveling pathogens — so far confined to science fiction stories — could in fact be powerful drivers of ecological change and threats to human health.

This technique could help an autonomous vehicle learn to compensate for slippery road conditions to avoid going into a skid, allow a robotic free-flyer to tow different objects in space, or enable a drone to closely follow a downhill skier despite being buffeted by strong winds.

The researchers’ approach incorporates certain structure from control theory into the process for learning a model in such a way that leads to an effective method of controlling complex dynamics, such as those caused by impacts of wind on the trajectory of a flying vehicle. One way to think about this structure is as a hint that can help guide how to control a system.

“The focus of our work is to learn intrinsic structure in the dynamics of the system that can be leveraged to design more effective, stabilizing controllers,” says Navid Azizan, the Esther and Harold E. Edgerton Assistant Professor in the MIT Department of Mechanical Engineering and the Institute for Data, Systems, and Society (IDSS), and a member of the Laboratory for Information and Decision Systems (LIDS). “By jointly learning the system’s dynamics and these unique control-oriented structures from data, we’re able to naturally create controllers that function much more effectively in the real world.”

Using this structure in a learned model, the researchers’ technique immediately extracts an effective controller from the model, as opposed to other machine-learning methods that require a controller to be derived or learned separately with additional steps. With this structure, their approach is also able to learn an effective controller using fewer data than other approaches. This could help their learning-based control system achieve better performance faster in rapidly changing environments.

“This work tries to strike a balance between identifying structure in your system and just learning a model from data,” says lead author Spencer M. Richards, a graduate student at Stanford University. “Our approach is inspired by how roboticists use physics to derive simpler models for robots. Physical analysis of these models often yields a useful structure for the purposes of control — one that you might miss if you just tried to naively fit a model to data. Instead, we try to identify similarly useful structure from data that indicates how to implement your control logic.”

Additional authors of the paper are Jean-Jacques Slotine, professor of mechanical engineering and of brain and cognitive sciences at MIT, and Marco Pavone, associate professor of aeronautics and astronautics at Stanford. The research will be presented at the International Conference on Machine Learning (ICML).

Learning a controller

Determining the best way to control a robot to accomplish a given task can be a difficult problem, even when researchers know how to model everything about the system.

A controller is the logic that enables a drone to follow a desired trajectory, for example. This controller would tell the drone how to adjust its rotor forces to compensate for the effect of winds that can knock it off a stable path to reach its goal.

This drone is a dynamical system — a physical system that evolves over time. In this case, its position and velocity change as it flies through the environment. If such a system is simple enough, engineers can derive a controller by hand.

Modeling a system by hand intrinsically captures a certain structure based on the physics of the system. For instance, if a robot were modeled manually using differential equations, these would capture the relationship between velocity, acceleration, and force. Acceleration is the rate of change in velocity over time, which is determined by the mass of and forces applied to the robot.

But often the system is too complex to be exactly modeled by hand. Aerodynamic effects, like the way swirling wind pushes a flying vehicle, are notoriously difficult to derive manually, Richards explains. Researchers would instead take measurements of the drone’s position, velocity, and rotor speeds over time, and use machine learning to fit a model of this dynamical system to the data. But these approaches typically don’t learn a control-based structure. This structure is useful in determining how to best set the rotor speeds to direct the motion of the drone over time.

Once they have modeled the dynamical system, many existing approaches also use data to learn a separate controller for the system.

“Other approaches that try to learn dynamics and a controller from data as separate entities are a bit detached philosophically from the way we normally do it for simpler systems. Our approach is more reminiscent of deriving models by hand from physics and linking that to control,” Richards says.

Identifying structure

The team from MIT and Stanford developed a technique that uses machine learning to learn the dynamics model, but in such a way that the model has some prescribed structure that is useful for controlling the system.

With this structure, they can extract a controller directly from the dynamics model, rather than using data to learn an entirely separate model for the controller.

“We found that beyond learning the dynamics, it’s also essential to learn the control-oriented structure that supports effective controller design. Our approach of learning state-dependent coefficient factorizations of the dynamics has outperformed the baselines in terms of data efficiency and tracking capability, proving to be successful in efficiently and effectively controlling the system’s trajectory,” Azizan says.

When they tested this approach, their controller closely followed desired trajectories, outpacing all the baseline methods. The controller extracted from their learned model nearly matched the performance of a ground-truth controller, which is built using the exact dynamics of the system.

“By making simpler assumptions, we got something that actually worked better than other complicated baseline approaches,” Richards adds.

The researchers also found that their method was data-efficient, which means it achieved high performance even with few data. For instance, it could effectively model a highly dynamic rotor-driven vehicle using only 100 data points. Methods that used multiple learned components saw their performance drop much faster with smaller datasets.

This efficiency could make their technique especially useful in situations where a drone or robot needs to learn quickly in rapidly changing conditions.

Plus, their approach is general and could be applied to many types of dynamical systems, from robotic arms to free-flying spacecraft operating in low-gravity environments.

In the future, the researchers are interested in developing models that are more physically interpretable, and that would be able to identify very specific information about a dynamical system, Richards says. This could lead to better-performing controllers.

This research is supported, in part, by the NASA University Leadership Initiative and the Natural Sciences and Engineering Research Council of Canada.

Researchers, including Jason Nesbitt, associate professor of archaeology at Tulane University School of Liberal Arts, performed genetic testing on individuals buried at Machu Picchu in order to learn more about the people who lived and worked there.

Machu Picchu is a UNESCO World Heritage Site located in the Cusco region of Peru. It is one of the most well-known archaeological sites in the world and attracts hundreds of thousands of visitors every year. It was once part of a royal estate of the Inca Empire.

Like other royal estates, Machu Picchu was home not only to royalty and other elite members of Inca society, but also to attendants and workers, many of whom lived in the estate year-round. These residents did not necessarily come from the local area, though it is only in this study that researchers have been able to confirm, with DNA evidence, the diversity of their backgrounds. “It’s telling us, not about elites and royalty, but lower status people,” Nesbitt said. “These were burials of the retainer population.”

This DNA analysis works in much the same way that modern genetic ancestry kits work. The researchers compared the DNA of 34 individuals buried at Machu Picchu to that of individuals from other places around the Inca Empire as well as some modern genomes from South America to see how closely related they might be.

The results of the DNA analysis showed that the individuals had come from throughout the Inca Empire, some as far away as Amazonia. Few of them had shared DNA with each other, showing that they had been brought to Machu Picchu as individuals rather than as part of a family or community group.

“Now, of course, genetics doesn’t translate into ethnicity or anything like that,” said Nesbitt of the results, “but that shows that they have distinct origins within different parts of the Inca Empire.”

“The study does really reinforce a lot of other types of research that have been done at Machu Picchu and other Inca sites,” Nesbitt said. The DNA analysis supports historical documentation and archaeological studies of the artifacts found associated with the burials.

This study is part of a larger movement in archaeology to combine traditional archaeological techniques with new technologies and scientific analyses. This combination of fields leads to a more complete understanding of the discoveries made.