Goto

Collaborating Authors

Case-Based Behavior Adaptation Using an Inverse Trust Metric

AAAI Conferences

Robots are added to human teams to increase the team's skills or capabilities but in order to get the full benefit the teams must trust the robots. We present an approach that allows a robot to estimate its trustworthiness and adapt its behavior accordingly. Additionally, the robot uses case-based reasoning to store previous behavior adaptations and uses this information to perform future adaptations. In a simulated robotics domain, we compare case-based behavior adaption to behavior adaptation that does not learn and show it significantly reduces the number of behaviors that need to be evaluated before a trustworthy behavior is found.


Adapting Autonomous Behavior Based on an Estimate of an Operator's Trust

AAAI Conferences

Robots can be added to human teams to provide improved capabilities or to perform tasks that humans are unsuited for. However, in order to get the full benefit of the robots the human teammates must use the robots in the appropriate situations. If the humans do not trust the robots, they may underutilize them or disuse them which could result in a failure to achieve team goals. We present a robot that is able to estimate its trustworthiness and adapt its behavior accordingly. This technique helps the robot remain trustworthy even when changes in context, task or teammates are possible.


Scale invariant robot behavior with fractals

#artificialintelligence

Robots deployed at orders of magnitude different size scales, and that retain the same desired behavior at any of those scales, would greatly expand the environments in which the robots could operate. However it is currently not known whether such robots exist, and, if they do, how to design them. Since self similar structures in nature often exhibit self similar behavior at different scales, we hypothesize that there may exist robot designs that have the same property. Here we demonstrate that this is indeed the case for some, but not all, modular soft robots: there are robot designs that exhibit a desired behavior at a small size scale, and if copies of that robot are attached together to realize the same design at higher scales, those larger robots exhibit similar behavior. We show how to find such designs in simulation using an evolutionary algorithm.


Predator-driven natural selection on risk-taking behavior in anole lizards

Science

Biologists have long debated the role of behavior in evolution, yet understanding of its role as a driver of adaptation is hampered by the scarcity of experimental studies of natural selection on behavior in nature. After showing that individual Anolis sagrei lizards vary consistently in risk-taking behaviors, we experimentally established populations on eight small islands either with or without Leiocephalus carinatus, a major ground predator. We found that selection predictably favors different risk-taking behaviors under different treatments: Exploratory behavior is favored in the absence of predators, whereas avoidance of the ground is favored in their presence. On predator islands, selection on behavior is stronger than selection on morphology, whereas the opposite holds on islands without predators. Our field experiment demonstrates that selection can shape behavioral traits, paving the way toward adaptation to varying environmental contexts.


Differential Replication in Machine Learning

arXiv.org Machine Learning

When deployed in the wild, machine learning models are usually confronted with data and requirements that constantly vary, either because of changes in the generating distribution or because external constraints change the environment where the model operates. To survive in such an ecosystem, machine learning models need to adapt to new conditions by evolving over time. The idea of model adaptability has been studied from different perspectives. In this paper, we propose a solution based on reusing the knowledge acquired by the already deployed machine learning models and leveraging it to train future generations. This is the idea behind differential replication of machine learning models. "If during the long course of ages and under varying conditions of life, organic beings vary at all in the several parts of their organization, [...] I think it would be a most extraordinary fact if no variation ever had occurred useful to each being's own welfare, in the same way as so many variations have occurred useful to man. But if variations useful to any organic being do occur, assuredly individuals thus characterized will have the best chance of being preserved in the struggle for life; and from the strong principle of inheritance they will tend to produce offspring similarly characterized. This principle of preservation, I have called, for the sake of brevity, Natural Selection."