Leading or Following? Dyadic Robot Imitative Interaction Using the Active Inference Framework

Wirkuttis, Nadine; Tani, Jun

doi:10.1109/lra.2021.3090015

Cited by 14 publications

(21 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This result of reduced flexibility under strong meta-prior condition is consistent with the finding reported by (Wirkuttis & Tani, 2021) that the PV-RNN with higher meta-prior had stronger intention and less flexible interaction with others because the top-down prior belief had more effects on generated behaviors than bottom-up sensory signals. In addition to reproducing this finding, we found that behavioral flexibility was improved by increasing stimulus noise under the strong meta-prior condition.…”

Section: Reduced Flexibility and Pathology Of Asdsupporting

confidence: 90%

“…Therefore, PV-RNN can be considered as a powerful tool for investigating Bayesian brain hypothesis. Indeed, PV-RNN was useful for modeling uncertainty estimations (Ahmadi & Tani, 2019), goal-oriented behavior (Matsumoto & Tani, 2020), sensory attenuation (Idei, Ohata, Yamashita, Ogata, & Tani, 2022), and social interaction (Ohata & Tani, 2020;Wirkuttis & Tani, 2021).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Comprehensive Computational Simulation Framework for Atypical Development using Hierarchical Bayesian Neural Network Model

Soda¹,

Ahmadi²,

Tani³

et al. 2023

Preprint

View full text Add to dashboard Cite

Investigating the pathological mechanisms of developmental disorders including autism spectrum disorder is a challenge, because of the fact that the presented symptoms are a result of complex and dynamic factors such as genes, molecules, neural networks, cognitive behavior, environment, and developmental learning. In recent years, computational methods, including the Bayesian brain hypothesis, have been expected to provide a unified framework for understanding developmental disorders that explain the neurocognitive functions by describing the interactions between these factors. However, this approach is still limited because most studies have focused on cross-sectional task performance and lacked the perspectives of developmental learning (i.e., acquisitions of knowledge reflecting probabilistic and hierarchical structures in the environment). In this study, we proposed a new research method for understanding the mechanisms of the acquisition and its failures in hierarchical Bayesian representations using a state-of-the-art computational model, referred to as ``in silico neurodevelopment framework for atypical representation learning.'' Using the proposed framework, simple simulation experiments were conducted to examine whether manipulating the neural stochasticity and noise levels in external environments can lead to the abnormal acquisition of hierarchical Bayesian representation and reduced flexibility. Consequently, networks with normal neural stochasticity were able to acquire hierarchical representations that reflected the underlying probabilistic structures in the environment, including higher-order representation, and exhibited good flexibility. When the neural stochasticity was high, top-down generation using higher-order representation was impaired, although the flexibility did not differ from that of the normal settings. On the other hand, the networks with extremely low neural stochasticity demonstrated reduced flexibility and abnormal hierarchical representation. However, this altered acquisition of higher-order representation and flexibility was ameliorated by increasing the level of noises in external stimuli. These results demonstrated that our proposed method is useful for investigating developmental disorders with bridging multiple factors including the inherent characteristics of the neural dynamics, acquisitions of hierarchical representation, and external environment.

show abstract

Section: Reduced Flexibility and Pathology Of Asdsupporting

confidence: 90%

Section: Introductionmentioning

confidence: 99%

Comprehensive Computational Simulation Framework for Atypical Development using Hierarchical Bayesian Neural Network Model

Soda¹,

Ahmadi²,

Tani³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…The majority of applications have focused on cognitive neuroscience, with a particular focus on modelling decision-making under uncertainty. Nonetheless, the framework has broad applicability and has recently been applied to diverse disciplines, ranging from computational models of psychopathology [5,6,7,8], control theory [9,10,11] and reinforcement learning [12,13,14,15,16], through to social cognition [17,18,19] and even real-world engineering problems [20,21,22]. While in recent years, some of the code arising from the active inference literature has been written in open source languages like Python and Julia [23,24,25,26,16], to-date, the most popular software for simulating active inference agents is the DEM toolbox of SPM [27,28].…”

Section: Statement Of Needmentioning

confidence: 99%

pymdp: A Python library for active inference in discrete state spaces

Heins,

Millidge,

Demekas

et al. 2022

Preprint

View full text Add to dashboard Cite

“…After the learning process has converged, action generation can be conducted. The basic architecture described above has been extended and implemented in various robotic experimental tasks including human-robot imitative interaction [84], [85], dyadic robot imitative interaction [86], goal-directed planning for robot object manipulation [87], and goal-directed planning using active inference and reinforcement learning in navigation [29].…”

Section: Hierarchical Representationmentioning

confidence: 99%

Active Inference in Robotics and Artificial Agents: Survey and Challenges

Lanillos¹,

Meo²,

Pezzato³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Active inference is a mathematical framework which originated in computational neuroscience as a theory of how the brain implements action, perception and learning. Recently, it has been shown to be a promising approach to the problems of state-estimation and control under uncertainty, as well as a foundation for the construction of goal-driven behaviours in robotics and artificial agents in general. Here, we review the state-of-the-art theory and implementations of active inference for state-estimation, control, planning and learning; describing current achievements with a particular focus on robotics. We showcase relevant experiments that illustrate its potential in terms of adaptation, generalization and robustness. Furthermore, we connect this approach with other frameworks and discuss its expected benefits and challenges: a unified framework with functional biological plausibility using variational Bayesian inference.

show abstract

Leading or Following? Dyadic Robot Imitative Interaction Using the Active Inference Framework

Cited by 14 publications

References 24 publications

Comprehensive Computational Simulation Framework for Atypical Development using Hierarchical Bayesian Neural Network Model

Comprehensive Computational Simulation Framework for Atypical Development using Hierarchical Bayesian Neural Network Model

pymdp: A Python library for active inference in discrete state spaces

Active Inference in Robotics and Artificial Agents: Survey and Challenges

Contact Info

Product

Resources

About