305 research outputs found
A Deep Learning-Based Emotion Recognition Pipeline for Public Speaking Anxiety Detection in Social Robotics
No full textSocial robots are increasingly employed as personalized coaches in educational settings, offering new opportunities for applications such as public speaking training. In this domain, emotional self-regulation plays a crucial role, especially for students presenting in a non-native language. This study proposes a novel pipeline for detecting public speaking anxiety (PSA) using multimodal emotion recognition. Unlike traditional datasets that typically rely on acted emotions, we consider spontaneous data from students interacting naturally with a social robot coach. Emotional labels are generated through knowledge distillation, enabling the creation of soft labels that reflect the emotional valence of each presentation. We introduce a lightweight multimodal model that integrates speech prosody and body posture to classify speakers by anxiety level, without relying on linguistic content. Evaluated on a collected dataset of student presentations, the system achieves 74.67% accuracy and an F1-score of 0.64. The model can operate completely disconnected from the transmission network on an NVIDIA Jetson board, safeguarding data privacy and demonstrating its feasibility for real-world deployment
Applications Of Automata Theory And Algebra : Via The Mathematical Theory of Complexity to Biology, Physics, Psychology, Philosophy, and Games
No full textThis book was originally written in 1969 by Berkeley mathematician John Rhodes. It is the founding work in what is now called algebraic engineering, an emerging field created by using the unifying scheme of finite state machine models and their complexity to tie together many fields: finite group theory, semigroup theory, automata and sequential machine theory, finite phase space physics, metabolic and evolutionary biology, epistemology, mathematical theory of psychoanalysis, philosophy, and game theory. The author thus introduced a completely original algebraic approach to complexity and the understanding of finite systems. The unpublished manuscript, often referred to as The Wild Book, became an underground classic, continually requested in manuscript form, and read by many leading researchers in mathematics, complex systems, artificial intelligence, and systems biology. Yet it has never been available in print until now. This first published edition has been edited and updated by Chrystopher Nehaniv for the 21st century. Its novel and rigorous development of the mathematical theory of complexity via algebraic automata theory reveals deep and unexpected connections between algebra (semigroups) and areas of science and engineering. Co-founded by John Rhodes and Kenneth Krohn in 1962, algebraic automata theory has grown into a vibrant area of research, including the complexity of automata, and semigroups and machines from an algebraic viewpoint, and which also touches on infinite groups, and other areas of algebra. This book sets the stage for the application of algebraic automata theory to areas outside mathematics. The material and references have been brought up to date by the editor as much as possible, yet the book retains its distinct character and the bold yet rigorous style of the author. Included are treatments of topics such as models of time as algebra via semigroup theory; evolution-complexity relations applicable to both ontogeny and evolution; an approach to classification of biological reactions and pathways; the relationships among coordinate systems, symmetry, and conservation principles in physics; discussion of punctuated equilibrium (prior to Stephen Jay Gould); games; and applications to psychology, psychoanalysis, epistemology, and the purpose of life. The approach and contents will be of interest to a variety of researchers and students in algebra as well as to the diverse, growing areas of applications of algebra in science and engineering. Moreover, many parts of the book will be intelligible to non-mathematicians, including students and experts from diverse backgrounds
Finite Automata Models: Algorithm, Application, and Semigroup Study
No full textA finite automaton consists of states and state transitions labeled by letters of a finite alphabet. Every letter describes a transformation of the automaton's state space, generating a corresponding transformation semigroup. A stochastic version of automata has a probability distribution over the alphabet at every state, that allows assigning the likelihoods to generated sequences of letters. In this thesis, we explore two models: deterministic probabilistic finite automaton (DPFA) considered from the practical side, and the sandpile model considered from the theoretical side.
For the first model, the presented study describes an algorithm for reconstructing DPFA from sequences of discrete observations using an n-gram merging method, including the practical implementation of the method, its experimental evaluation on case studies, and the application of this automata-based technique to the neurobiological data. For considered examples the performance is compared to the causal state splitting reconstruction (CSSR) technique: both methods achieve high quality in approximating the probability distribution over strings. Considering if transformation semigroups of reconstructed automata contain subgroups corresponding to those in examples, CSSR shows a good result for preserving cyclic permutation groups, but not the n-gram merging method, whose transformation semigroup is generally aperiodic. The application considered in this study uses electroencephalographic (EEG) microstate sequences and considers the question of distinguishing the participant groups (meditators and controls) and cognitive modes (mind-wandering, verbalization, visualization) by separating DPFA machines inferred from EEG data in a metric space. The separation between participant groups is achieved for many parameter settings with linear criterion (requiring non-overlapping clusters) and for a few instances with strict criterion (requiring dense distant clusters). Both criteria show great reliability when validated using permuted data. The separation of cognitive modes only demonstrated partial success with noticeably better performance within the group of controls and more instances of separation corresponding to the visualization condition.
For the second, sandpile model, the presented study concentrates on the properties of their transformation semigroups for the standard Abelian sandpiles on circle graphs and the modified model, non-Abelian sandpiles on rooted trees. The exploration addresses the structure of recurrent configurations, the wreath product decomposition of semigroups, and the decomposition-based complexity measure. The identity and generator configurations of the recurrent group of Abelian sandpiles on circles are described, giving an explicit alternative way of understanding their well-known cyclic structure. The complexity of arbitrary finite Abelian semigroup is shown to be at most one. The embedding of the sandpile semigroup into the wreath product of flip-flop semigroups is constructed for non-Abelian sandpiles on rooted trees, implying its aperiodic complexity is the depth of the tree
Synthesis of User Interfaces with Categorical Methods
No full textWe apply category theory to modeling user interfaces, focusing on the interaction between functional configuration and user perception. By representing user interfaces as directed labeled multigraphs and applying pullback constructions to the category of directed labeled multigraphs, the study formalizes interface structures in a way that encompasses both technical operations and users' perceptual capabilities. Users interact with interfaces by identifying affordances that hint at possible actions. We propose a perception-driven interpretation of the interface as a set of affordances available to the user.
Conversely, the structure of the interface is formed by its functional components and presented in different states. This double strategy aims to provide a tool for studying the usability and unambiguity of interactive systems and analyzing how interfaces communicate functionality to users through structural design.
A major component of this research is the study of user profiles, which define the relationship between the interface and the cognitive and physical characteristics of users. Profiles are encoded in the interface representation by graphs as certain filters of the pullback graph to match the human perception. This approach provides a basis for assessing usability for people with different cognitive and physical abilities.
Ambiguity is the presence of multiple possible meanings, interpretations, or outcomes that can cause uncertainty. It affects the user's sense of control over the interface. The thesis examines ambiguity in user interfaces within a categorical formal representation. By applying the conditions to the pullback, the study proposes a formal method to detect the ambiguity of the interface from the perspective of a particular user profile, supported by provided case studies
An Algebraic Perspective on Game Theory
No full textEven though the fields of game theory and algebra are well-established, limited literature analyzes games using modern algebraic theories. In 2009, Rhodes published a game complexity study titled \textit{Complexity of Games} via \textit{Applications of Automata Theory and Algebra}, which uses Krohn-Rhodes theory to analyze the best-play automata of the two-player complete-information sequential games. It defines the key terminologies, derives the automata representations from complete game trees, and proposes several theorems and conjectures. Rhodes' work heavily inspires this thesis as we expand on the established theoretical frameworks and challenge some of Rhodes' claims.
We begin by presenting the essential terminologies of the semigroup theory and Krohn-Rhodes theory. We then examine the existing methods for complexity analysis and propose novel algorithms to calculate the complexity lower bound for transformation semigroups. These algorithms underwent multiple iterations of improvements and contributed considerably to the analysis in the subsequent chapters.
A major component of the thesis is a direct expansion of Rhodes' work with respect to the two-player complete-information sequential games. We represent the gameplay machines identical to Rhodes'; however, we also consider symmetries in the structures of games to reduce the state space. We then developed practical methodologies to evaluate the algebraic structure of several simple games, such as Tic-Tac-Toe and used them to verify some of Rhodes' conjectures.
Another major component of the thesis is the algebraic analysis of extensive-form games, which encompass a much broader category of games (including the complete-information sequential games). This exploration encountered many challenges, in particular, the analysis of nondeterminism associated with mixed strategy profiles. We proposed the usage of classic powerset construction of the nondeterministic finite automata, but it turned out to be highly problematic due to state explosion and the possibility of collapsing into a single `superset' state. To overcome these issues, we proposed an alternative construction that captures the nondeterminism without the drawbacks of the powerset construction. This construction enables us to analyze the algebraic structure and complexity of games effectively.
Overall, this thesis serves to bridge the gap between the classical game theory and the algebraic automata theory
Robots Learning to Say `No': Prohibition and Rejective Mechanisms in Acquisition of Linguistic Negation
No full text`No' belongs to the first ten words used by children and embodies the first
active form of linguistic negation. Despite its early occurrence the details of
its acquisition process remain largely unknown. The circumstance that `no'
cannot be construed as a label for perceptible objects or events puts it
outside of the scope of most modern accounts of language acquisition. Moreover,
most symbol grounding architectures will struggle to ground the word due to its
non-referential character. In an experimental study involving the child-like
humanoid robot iCub that was designed to illuminate the acquisition process of
negation words, the robot is deployed in several rounds of speech-wise
unconstrained interaction with na\"ive participants acting as its language
teachers. The results corroborate the hypothesis that affect or volition plays
a pivotal role in the socially distributed acquisition process. Negation words
are prosodically salient within prohibitive utterances and negative intent
interpretations such that they can be easily isolated from the teacher's speech
signal. These words subsequently may be grounded in negative affective states.
However, observations of the nature of prohibitive acts and the temporal
relationships between its linguistic and extra-linguistic components raise
serious questions over the suitability of Hebbian-type algorithms for language
grounding.Comment: Submitted journal article. 21 pages main paper plus 28 pages
supplementary information / appendix. 8 figures in main pape
Current work and open problems:: A road-map for research into the emergence of communication and language
No full textThis book brings together work on the emergence of communication and language from researchers working in a broad array of scientific paradigms in North America, Europe, Japan and Africa. © 2007 Springer-Verlag London
Engineering Social Learning Mechanisms for Multi-Agent Interaction
No full textThis thesis is strongly inspired by literature on animal social learning, applying it to multi-robot as well as human-robot interaction scenarios, Social learning, which can include complex or simple social mechanisms, allow us to understand cooperation and communication in animals, giving them better chances to survive for longer and thrive as a society. For this dissertation, to translate this understanding into socially rich behavior among multi-agent robots and Human-Robot Interaction, two experiments were conducted.
The first experiment focused on how social learning might optimize cooperation among robots (in a robot 'society') for the problem of foraging. The task utilizes small and simple swarm robots to understand how such social mechanisms might play a role in establishing rules for emergent group behavior and how social rules might be engineered to gain useful effects in a group of robots. The study investigated exploratory behavior without interaction (asocial) and with interaction (social). The results from this exploratory study suggest that deterministic asocial exploration is best performed by a Spiral exploration mechanisms. However, these asocial exploration strategies are eclipsed by certain types of social reward sharing strategies as long as sharing occurs for at least half the lifetime of the robots. Sharing locations of reward caches for all time is of course the most optimal, but comes at the cost of communicating longer and hence using more energy both on the sender and receiver’s end. An analysis of a compromise strategy between completely asocial exploration and social reward location sharing is performed using strategies termed critical and conditional learning. It is found that the number of reward caches located through critical and conditional learning are intermediary to the two extremes, namely completely asocial and completely social foraging.
The second experiment sought to understand if and how other types of social learning mechanisms such as observational conditioning can facilitate social information spread to human participants. The question of whether, and to what extent, a robot can influence a human's actions is asked through a study designed to understand if emotions displayed by a robot demonstrators can influence human observers. An immersive first-person gaming experience utilizing Unity was designed where a robot demonstrator reacted either positively or negatively to an external stimulus. Objective (position of player in-game) and subjective (Questionnaire) data collected on the human participants' reactions suggests that the virtual robot agent is successful in socially transmitting information.
Through these studies, I seek to contribute to the understanding of the role simple social learning mechanisms can play in information transfer among human and robot agents, and to identify useful metrics for the detection of such social mechanisms
Multiple motivations for imitation in infancy
No full textAs the chapters in this book attest, during the last decade the study of imitation has become a topic of central importance through a diverse range of disciplines. There have been a greater number of controlled studies of imitation in non-human animals than ever before. Possible neural foundations of imitation have been identified. Encouraging progress has been made in developing imitation in constructed systems. Our understanding of the processes and mechanisms of imitation has been markedly advanced. Nonetheless, during this period most researchers have primarily focused on how imitation facilitates the acquisition of new skills or behaviours. In so doing, some important aspects of imitation have been neglected. In human development, infants imitate for a wide variety of reasons, both within and across different developmental stages and within and across different contexts. More specifically, for human infants imitation is an important form of pre-verbal communication that provides a means by which they can engage in social interaction. Our aim in this chapter is to provide an overview of the evidence that infants imitate not only to acquire new skills but also to engage socially with others, and this social engagement can itself take a number of different forms, with imitation being used flexibly as a means to various social ends. Before going further a brief note on definition is warranted
- …
