Eleni Petraki

Trusted Autonomy and Cognitive Cyber Symbiosis: Open Challenges

Abstract This paper considers two emerging interdisciplinary, but related topics that are likely to create tipping points in advancing the engineering and science areas. Trusted Autonomy (TA) is a field of research that focuses on understanding and designing the interaction space between two entities each of which exhibits a level of autonomy. These entities can be humans, machines, or a mix of the two. Cognitive Cyber Symbiosis (CoCyS) is a cloud that uses humans and machines for decision-making. In CoCyS, human–machine teams are viewed as a network with each node comprising humans (as computational machines) or computers. CoCyS focuses on the architecture and interface of a Trusted Autonomous System. This paper examines these two concepts and seeks to remove ambiguity by introducing formal definitions for these concepts. It then discusses open challenges for TA and CoCyS, that is, whether a team made of humans and machines can work in fluid, seamless harmony.

The

Trust is a fundamental concept that underpins the coherence and resilience of social systems and shapes human behavior. Despite the importance of trust as a social and psychological concept, the concept has not gained much attention from evolutionary game theorists. In this letter, an N-player trust-based social dilemma game is introduced. While the theory shows that a society with no untrustworthy individuals would yield maximum wealth to both the society as a whole and the individuals in the long run, evolutionary dynamics show this ideal situation is reached only in a special case when the initial population contains no untrustworthy individuals. When the initial population consists of even the slightest number of untrustworthy individuals, the society converges to zero trusters, with many untrustworthy individuals. The promotion of trust is an uneasy task, despite the fact that a combination of trusters and trustworthy trustees is the most rational and optimal social state. This letter presents the game and results of replicator dynamics in a hope that researchers in evolutionary games see opportunities in filling this critical gap in the literature.

N

The concept of trust has attracted the attention of many researchers over the years who studied the impact of trust in many domains. Trust is a ubiquitous concept. It is pervasive in every aspect of our life, from interpersonal relationships to national defence and security applications. However, despite the vast literature on trust, we are not close enough to mastering the dynamics of trust. One reason is that if we define procedural steps for trust, we simultaneously define steps for deception; thus, we simply define a vacuous cycle. Another reason is that, the dynamics of trust change as the world changes. But how can we then study trust? This paper connects the interdisciplinary literature to synthesize a Computational Red Teaming (CRT) based model of trust that defines opportunities whereby computational intelligence techniques, more specifically, evolutionary game theory researchers, can contribute to this vastly growing research area. We offer a position on the topic by reviewing games for trust and introduce a new theoretic game to study influence and transfer of trust.

-Player Trust Game and its Replicator Dynamics

Translator Stylometry is a small but growing area of research in computational linguistics. Despite the research proliferation on the wider research field of authorship attribution using computational linguistics techniques, the translator stylometry problem is more challenging and there is no sufficient literature on the topic. Some authors even claimed that this problem does not have a solution; a claim we will challenge in this paper. We present an innovative set of translator stylometric features that can be used as signatures to detect and identify translators. The features are based on the concept of network motifs: small graph local substructures which have been used successfully in characterizing global network dynamics. The text is transformed into a network, where words become nodes and their adjacencies in a sentence are represented through links. Motifs of size 3 are then extracted from this network and their distribution is used as a signature for the corresponding translator. We then investigate the impact of sample size, method of normalization and imbalance dataset on classification accuracy. We also adopt the Fuzzy Lattice Reasoning Classifier (FLR) among others, where FLR achieved the best performance with a classification accuracy reaching the 70% mark.

On trust and influence: A computational red teaming game theoretic perspective

The use of evolutionary computation to automatically narrate a story in a natural language, such as English, is a very daunting task. Two main challenges are addressed in this paper. First, how to represent a story in a form that is simple for evolution to work on? Second, how to evaluate stories using proper objective metrics? We address the first challenge by introducing a permutation-based linear representation that relies on capturing the events in a story in a genome, and on transforming any sequence represented by this genome into a valid story using a genotype-phenotype mapping. This mapping uses causal relationships in a story as constraints. The second challenge is addressed by conducting human-based experiments to collect subjective measurements of two categories of familiar and unfamiliar stories to the participants. The data collected from this exercise are then correlated with objective metrics that we designed to capture the quality of a story. Results reveal interesting relationships that are discussed in details in the paper.

A computational linguistic approach for the identification of translator stylometry using Arabic-English text

In this article, we present a new type of classification problem, which we call Comparative Classification Problem (CCP), where we use the term data record to refer to a block of instances. Given a single data record with n instances for n classes, the CCP problem is to map each instance to a unique class. This problem occurs in a wide range of applications where the independent and identically distributed assumption is broken down. The primary difference between CCP and classical classification is that in the latter, the assignment of a translator to one record is independent of the assignment of a translator to a different record. In CCP, however, the assignment of a translator to one record within a block excludes this translator from further assignments to any other record in that block. The interdependency in the data poses challenges for techniques relying on the independent and identically distributed (iid) assumption. In the Pairwise CCP (PWCCP), a pair of records is grouped together. The key difference between PWCCP and classical binary classification problems is that hidden patterns can only be unmasked by comparing the instances as pairs. In this article, we introduce a new algorithm, PWC4.5, which is based on C4.5, to manage PWCCP. We first show that a simple transformation—that we call Gradient-Based Transformation (GBT)—can fix the problem of iid in C4.5. We then evaluate PWC4.5 using two real-world corpora to distinguish between translators on Arabic-English and French-English translations. While the traditional C4.5 failed to distinguish between different translators, GBT demonstrated better performance. Meanwhile, PWC4.5 consistently provided the best results over C4.5 and GBT.

From subjective to objective metrics for evolutionary story narration using event permutations

Our previous work introduced the N player trust game and examined the dynamics of this game using replicator dynamics for an infinite population. In finite populations, quantization becomes a necessity that introduces discontinuity in the trajectory space, which can impact the dynamics of the game differently. In this paper, we present an analysis of replicator dynamics of the N player trust game in finite populations. The analysis reveals that, quantization indeed introduces fixed points in the interior of the 2-simplex that were not present in the infinite population analysis. However, there is no guarantee that these fixed points will continue to exist for any arbitrary population size; thus, they are clearly an artifact of quantization. In general, the evolutionary dynamics of the finite population are qualitatively similar to the infinite population. This suggests that for the proposed trust game, trusters will be extinct if the population contains an untrustworthy player. Therefore, trusting is an evolutionary unstable strategy.

Pairwise Comparative Classification for Translator Stylometric Analysis

Communication has been an active field of research in Robotics. However, less work has been done in the ability of robots to negotiate meanings of the world through storytelling. In this paper, we address this gap from the perspective of evolving stories. By approximating human evaluation of stories to guide the evolution, we can automate the story evolutionary process without interacting with humans. First, a multi-objective story evolution approach is applied where the approximated human story evaluation model automatically evaluates the subjective story metrics such as coherence, novelty and interestingness. We then use humans again to validate the stories narrated by the machine. Results show that for each of the human subjects, the stories collected after story evolution are regarded as better stories compared to the initial stories. Some interesting relationships are revealed and discussed in details.

Finite Population Trust Game Replicators

Stories are useful tools with which we can exchange experience learnt in social contexts, ways to communicate futures in strategic planning, and unique building blocks that connect meanings in a movie or a virtual environment. Evolutionary computation (EC) techniques have the potential to overcome existing limitations in automated storytelling, whereby evolution can provide a process of innovation. However, one source of complexity lies in the transformation of a story in a natural language into a representation that EC can evolve easily. Another complexity arises from the fact that the ultimate judge for the quality of a story is a human being, and humans are diverse in their taste. This paper attempts to tackle the above complexities through an automatic story narration application. We present a methodology which can transform a story written in English into an event-level and hierarchical-level grammar using a network representation. This approach makes it possible to devise an encoding scheme that translates a story narration with flashback into a chromosome and vice versa. We then discuss different metrics for the evolutionary narration problem and use 42 human participants to evaluate the generated narrations. To incorporate diversified human opinions, we propose to build individual human-surrogate models from the human-evaluation experiment and further fuse them into an ensemble. The ensembles of human surrogate models serve as the objective functions of multi-objective EC to guide the generation of desirable stories from human perspectives. We demonstrate that this approach is successful in evolving better narrations as assessed by 31 human participants.

Evolving Story Narrative using Surrogate Models of Human Judgement

Social dilemmas require individuals to tradeoff self interests against group interests. Considerable research effort has attempted to identify conditions that promote cooperation in these social dilemmas. It has previously been shown altruistic punishment can help promote cooperation but the mechanisms that make it work are not well understood. We have designed a multi-agent system to investigate altruistic punishment in tragedy of the commons social dilemmas. Players develop emotional responses as they interact with others. A zero order Seguno fuzzy system is used to model the player emotional responses. Players change strategies when their emotional level exceeds a personal emotional threshold. Trustworthiness of how other players will act in the future helps choose the new strategy. Our results show how strategies evolve in a finite population match predictions made using discrete replicator equations.