ABSTRACT
The behavior management controls proposed for social robots are mostly designed for highly controlled scenarios. In the real world though, robots have to adapt to new situations, generalizing learned behaviors. To address this adaptation challenge, neural network models with embedding layers could be used. We present here an approach to better understand the inductive biases of our robotic gaze model. It was trained with multimodal features as inputs -- either endogenous or exogenous to the robot. Inductive biases were explored by observing feature representations in the embedding spaces. We found that the model was able to distinguish between the robot speech intentions that either request or provide information. Similarly, pairs of partners seem grouped according to their social behavior (speaking time, gaze). Finally, we checked that these groupings had a real impact on the model's performance. Driving these biases when facing new people should allow to generate adapted behavior.
Supplemental Material
- Henny Admoni and Brian Scassellati. 2017. Social Eye Gaze in Human-Robot Interaction: A Review. J. Hum.-Robot Interact., Vol. 6, 1 (may 2017), 25--63.Google ScholarDigital Library
- Remi Cambuzat, Frédéric Elisei, Gérard Bailly, Olivier Simonin, and Anne Spalanzani. 2018. Immersive Teleoperation of the Eye Gaze of Social Robots Assessing Gaze-Contingent Control of Vergence, Yaw and Pitch of Robotic Eyes. In ISR 2018 - 50th International Symposium on Robotics. VDE, Munich, Germany, 232--239.Google Scholar
- Mireille Fares, Catherine Pelachaud, and Nicolas Obin. 2023. Zero-shot style transfer for gesture animation driven by text and speech using adversarial disentanglement of multimodal style encoding. Frontiers in Artificial Intelligence, Vol. 6 (2023), 1142997.Google ScholarCross Ref
- Laurent Prévot Frédéric Elisei, Gérard Bailly. 2023. RoboTrio2. https://hdl.handle.net/11403/robotrio/v2 ORTOLANG (Open Resources and TOols for LANGuage) textendash www.ortolang.fr.Google Scholar
- Léa Haefflinger, Frédéric Elisei, Silvain Gerber, Béatrice Bouchot, Jean-Philippe Vigne, and Gérard Bailly. 2023. On the Benefit of Independent Control of Head and Eye Movements of a Social Robot for Multiparty Human-Robot Interaction. In Human-Computer Interaction, Masaaki Kurosu and Ayako Hashizume (Eds.). Springer Nature Switzerland, Cham, 450--466.Google Scholar
- Thilo Hagendorff and Sarah Fabi. 2023. Why we need biased ai: How including cognitive biases can enhance ai systems. Journal of Experimental & Theoretical Artificial Intelligence (2023), 1--14.Google Scholar
- Eyke Hüllermeier, Thomas Fober, and Marco Mernberger. 2013. Inductive Bias. Springer New York, New York, NY, 1018--1018.Google Scholar
- Carlos Toshinori Ishi and Taiken Shintani. 2021. Analysis of Eye Gaze Reasons and Gaze Aversions During Three-Party Conversations. In Proc. Interspeech 2021. 1972--1976.Google ScholarCross Ref
- Wolfgang Kabsch. 1976. A solution for the best rotation to relate two sets of vectors. Acta Crystallographica Section A: Crystal Physics, Diffraction, Theoretical and General Crystallography, Vol. 32, 5 (1976), 922--923.Google ScholarCross Ref
- Diederik Kingma and Jimmy Ba. 2014. Adam: A Method for Stochastic Optimization. International Conference on Learning Representations (2014).Google Scholar
- Chinmaya Mishra and Gabriel Skantze. 2022. Knowing Where to Look: A Planning-based Architecture to Automate the Gaze Behavior of Social Robots. In 2022 31st IEEE International Conference on Robot and Human Interactive Communication (RO-MAN). 1201--1208.Google ScholarDigital Library
- Bilge Mutlu, Jodi Forlizzi, and Jessica Hodgins. 2006. A storytelling robot: Modeling and evaluation of human-like gaze behavior. In 2006 6th IEEE-RAS International Conference on Humanoid Robots. IEEE, 518--523.Google ScholarCross Ref
- Bilge Mutlu, Takayuki Kanda, Jodi Forlizzi, Jessica Hodgins, and Hiroshi Ishiguro. 2012. Conversational gaze mechanisms for humanlike robots. ACM Transactions on Interactive Intelligent Systems, Vol. 1 (01 2012), 12.Google ScholarDigital Library
- Leann Myers and Maria J Sirois. 2004. Spearman correlation coefficients, differences between. Encyclopedia of statistical sciences, Vol. 12 (2004).Google Scholar
- Yukiko I. Nakano, Takashi Yoshino, Misato Yatsushiro, and Yutaka Takase. 2015. Generating Robot Gaze on the Basis of Participation Roles and Dominance Estimation in Multiparty Interaction. ACM Trans. Interact. Intell. Syst., Vol. 5, 4, Article 22 (dec 2015), 23 pages.Google Scholar
- Alberto Parmiggiani, Marco Randazzo, Marco Maggiali, Frederic Elisei, Gerard Bailly, and Giorgio Metta. 2014. An articulated talking face for the iCub. In 2014 IEEE-RAS International Conference on Humanoid Robots. 1--6. https://doi.org/10.1109/HUMANOIDS.2014.7041309Google ScholarCross Ref
- Taiken Shintani, Carlos T. Ishi, and Hiroshi Ishiguro. 2021. Analysis of Role-Based Gaze Behaviors and Gaze Aversions, and Implementation of Robot's Gaze Control for Multi-Party Dialogue. In Proceedings of the 9th International Conference on Human-Agent Interaction (Virtual Event, Japan) (HAI '21). Association for Computing Machinery, New York, NY, USA, 332--336.Google ScholarDigital Library
- Gabriel Skantze. 2017. Predicting and Regulating Participation Equality in Human-Robot Conversations: Effects of Age and Gender. In Proceedings of the 2017 ACM/IEEE International Conference on Human-Robot Interaction (Vienna, Austria) (HRI '17). Association for Computing Machinery, New York, NY, USA, 196--204.Google ScholarDigital Library
- Shirui Wang, Wenan Zhou, and Chao Jiang. 2020. A survey of word embeddings based on deep learning. Computing, Vol. 102 (2020), 717--740.Google ScholarCross Ref
- Svante Wold, Kim Esbensen, and Paul Geladi. 1987. Principal component analysis. Chemometrics and intelligent laboratory systems, Vol. 2, 1--3 (1987), 37--52.Google Scholar
Index Terms
- Probing the Inductive Biases of a Gaze Model for Multi-party Interaction
Recommendations
Lexical Entrainment in Multi-party Human–Robot Interaction
Social RoboticsAbstractThis paper reports lexical entrainment in a multi-party human–robot interaction, wherein one robot and two humans serve as participants. Humans tend to use the same terms as their interlocutors while making conversation. This phenomenon is called ...
Conversational gaze mechanisms for humanlike robots
During conversations, speakers employ a number of verbal and nonverbal mechanisms to establish who participates in the conversation, when, and in what capacity. Gaze cues and mechanisms are particularly instrumental in establishing the participant roles ...
A gesture-centric Android system for multi-party human-robot interaction
Special Issue on HRI System StudiesNatural body gesturing and speech dialogue, is crucial for human-robot interaction (HRI) and human-robot symbiosis. Real interaction is not only with one-to-one communication but also among multiple people. We have therefore developed a system that can ...
Comments