Optimizing group learning: An evolutionary computing approach

Modeling Misinformation Spread in a Bounded Confidence Model: A Simulation Study

Misinformation has posed significant threats to all aspects of people's lives. One of the most active areas of research in misinformation examines how individuals are misinformed. In this paper, we study how and to what extent agents are misinformed in an extended bounded confidence model, which consists of three parts: (i) online selective neighbors whose opinions differ from their own but not by more than a certain confidence level; (ii) offline neighbors, in a Watts-Strogatz small-world network, whom an agent has to communicate with even though their opinions are far different from their own; and (iii) a Bayesian analysis. Furthermore, we introduce two types of epistemically irresponsible agents: agents who hide their honest opinions and focus on disseminating misinformation and agents who ignore the messages received and follow the crowd mindlessly. Simulations show that, in an environment with only online selective neighbors, the misinforming is more successful with broader confidence intervals. Having offline neighbors contributes to being cautious of misinformation, while employing a Bayesian analysis helps in discovering the truth. Moreover, the agents who are only willing to listen to the majority, regardless of the truth, unwittingly help to bring about the success of misinformation attempts, and they themselves are, of course, misled to a greater extent.

Network effects in a bounded confidence model

The bounded confidence model has become a popular tool for studying communities of epistemically interacting agents. The model makes the idealizing assumption that all agents always have access to all other agents' belief states. We draw on resources from network epistemology to do away with this assumption. In the model to be proposed, we impose an explicit communication network on a community, due to which each agent has access to the beliefs of only a selection of other agents. A much-discussed result from network epistemology shows that densely connected communication networks are not always preferable to sparser networks. The aim of this paper is to investigate whether there are any noteworthy network effects in a version of the bounded confidence model augmented with communication networks, and in particular whether the aforementioned result from network epistemology can be replicated in that version.

How explanation guides belief change

Philosophers have argued that people ought to change their graded beliefs via Bayes' rule. Recent work in psychology indicates that people sometimes violate that rule by attending to explanatory factors. Results from computational modeling suggest that such violations may actually be rational.

Collective preference learning in the best-of-n problem

Decentralised autonomous systems rely on distributed learning to make decisions and to collaborate in pursuit of a shared objective. For example, in swarm robotics the best-of-n problem is a well-known collective decision-making problem in which agents attempt to learn the best option out of n possible alternatives based on local feedback from the environment. This typically involves gathering information about all n alternatives while then systematically discarding information about all but the best option. However, for applications such as search and rescue in which learning the ranking of options is useful or crucial, best-of-n decision-making can be wasteful and costly. Instead, we investigate a more general distributed learning process in which agents learn a preference ordering over all of the n options. More specifically, we introduce a distributed rank learning algorithm based on three-valued logic. We then use agent-based simulation experiments to demonstrate the effectiveness of this model. In this context, we show that a population of agents are able to learn a total ordering over the n options and furthermore the learning process is robust to evidential noise. To demonstrate the practicality of our model, we restrict the communication bandwidth between the agents and show that this model is also robust to limited communications whilst outperforming a comparable probabilistic model under the same communication conditions.

Negative updating applied to the best-of-n problem with noisy qualities

The ability to perform well in the presence of noise is an important consideration when evaluating the effectiveness of a collective decision-making framework. Any system deployed for real-world applications will have to perform well in complex and uncertain environments, and a component of this is the limited reliability and accuracy of evidence sources. In particular, in swarm robotics there is an emphasis on small and inexpensive robots which are often equipped with low-cost sensors more prone to suffer from noisy readings. This paper presents an exploratory investigation into the robustness of a negative updating approach to the best-of-n problem which utilises negative feedback from direct pairwise comparison of options and opinion pooling. A site selection task is conducted with a small-scale swarm of five e-puck robots choosing between $$n=7$$ n = 7 options in a semi-virtual environment with varying levels of sensor noise. Simulation experiments are then used to investigate the scalability of the approach. We now vary the swarm size and observe the behaviour as the number of options n increases for different error levels with different pooling regimes. Preliminary results suggest that the approach is robust to noise in the form of noisy sensor readings for even small populations by supporting self-correction within the population.

The ecological rationality of explanatory reasoning

There is growing evidence that explanatory considerations influence how people change their degrees of belief in light of new information. Recent studies indicate that this influence is systematic and may result from people's following a probabilistic update rule. While formally very similar to Bayes' rule, the rule or rules people appear to follow are different from, and inconsistent with, that better-known update rule. This raises the question of the normative status of those updating procedures. Is the role explanation plays in people's updating their degrees of belief a bias? Or are people right to update on the basis of explanatory considerations, in that this offers benefits that could not be had otherwise? Various philosophers have argued that any reasoning at deviance with Bayesian principles is to be rejected, and so explanatory reasoning, insofar as it deviates from Bayes' rule, can only be fallacious. We challenge this claim by showing how the kind of explanation-based update rules to which people seem to adhere make it easier to strike the best balance between being fast learners and being accurate learners. Borrowing from the literature on ecological rationality, we argue that what counts as the best balance is intrinsically context-sensitive, and that a main advantage of explanatory update rules is that, unlike Bayes' rule, they have an adjustable parameter which can be fine-tuned per context. The main methodology to be used is agent-based optimization, which also allows us to take an evolutionary perspective on explanatory reasoning.