Search

Social Robotics and Human-Robot Interaction (HRI) research relies on different Affective Computing (AC) solutions for sensing, perceiving and understanding human affective behaviour during interactions. This may include utilising off-the-shelf affect perception models that are pre-trained on popular affect recognition benchmarks and directly applied to situated interactions. However, the conditions in situated human-robot interactions differ significantly from the training data and settings of these models. Thus, there is a need to deepen our understanding of how AC solutions can be best leveraged, customised and applied for situated HRI. This paper, while critiquing the existing practices, presents four critical lessons to be noted by the hitchhiker when applying AC for HRI research. These lessons conclude that: (i) The six basic emotions categories are irrelevant in situated interactions, (ii) Affect recognition accuracy (%) improvements are unimportant, (iii) Affect recognition does not generalise across contexts, and (iv) Affect recognition alone is insufficient for adaptation and personalisation. By describing the background and the context for each lesson, and demonstrating how these lessons have been learnt, this paper aims to enable the hitchhiker to successfully and insightfully leverage AC solutions for advancing HRI research., Comment: 11 pages, 3 figures, 1 table

As Facial Expression Recognition (FER) systems become integrated into our daily lives, these systems need to prioritise making fair decisions instead of aiming at higher individual accuracy scores. Ranging from surveillance systems to diagnosing mental and emotional health conditions of individuals, these systems need to balance the accuracy vs fairness trade-off to make decisions that do not unjustly discriminate against specific under-represented demographic groups. Identifying bias as a critical problem in facial analysis systems, different methods have been proposed that aim to mitigate bias both at data and algorithmic levels. In this work, we propose the novel usage of Continual Learning (CL), in particular, using Domain-Incremental Learning (Domain-IL) settings, as a potent bias mitigation method to enhance the fairness of FER systems while guarding against biases arising from skewed data distributions. We compare different non-CL-based and CL-based methods for their classification accuracy and fairness scores on expression recognition and Action Unit (AU) detection tasks using two popular benchmarks, the RAF-DB and BP4D datasets, respectively. Our experimental results show that CL-based methods, on average, outperform other popular bias mitigation techniques on both accuracy and fairness metrics.

Recent advancements in Artificial Intelligence (AI) and Human-Robot Interaction (HRI) have enabled robots to be integrated into daily human life. Operating in human-centred environments, these robots need to actively participate in the human ‘affective loop’, sensing and interpreting human socio-emotional behaviours while also learning to respond in a manner that fosters their social and emotional wellbeing. Embedding affective robots with learning mechanisms that enable such a robust understanding of human behaviour as well as their own role in an interaction forms the central focus of affective robotics research. Current Machine Learning (ML)-based solutions for realising affective capabilities in robots, be it robust affect perception or behaviour generation, are primed towards generalisation of application. Pre-trained on volumes of data, these solutions, although enabling a wide variety of applications, are static and unable to adapt sufficiently to the dynamics of real-world interactions. Affective robots, on the other hand, need personalised interaction capabilities that, sensitive to an individual’s socio-emotional behaviour, can adapt affective interactions towards them, expanding their learning on the go to include novel information, while ensuring past knowledge is preserved. Addressing these challenges, this dissertation proposes the novel application of the Continual Learning (CL) paradigm for affective robotics, enabling continual and lifelong adaptation capabilities in robots. It provides the foundational formulations that translate key principles of CL-based adaptation for affective learning in robots. Furthermore, investigating learning at every stage of the ‘affective loop’, it reflects upon the key desiderata for affective robots in terms of continual and personalised affect perception and context-appropriate behaviour generation. Starting with affect perception, this dissertation presents the first extensive benchmark on Continual Facial Expression Recognition (ConFER), evaluating CL-based approaches for continually learning facial expression classes under different learning settings. Despite enabling incremental learning, ConFER does not focus on personalised affect perception, another key requirement for affective robots. To address this, a novel framework is proposed using Continual Learning with Imagination for Facial Expression Recognition (CLIFER). Inspired from the cognitive processes in the human brain focused on memory-based learning and mental imagery, CLIFER incrementally learns facial expression classes while personalising towards individual affective expression, using imagination to augment person-specific learning. CLIFER is shown to achieve state-of-the-art (SOTA) results on different benchmark evaluations. Learning under such dynamic conditions, affective robots need to remain fair and equitable, ensuring no individual (or group) is disadvantaged and the robots’ perception is bias-free. Exploring different domain groups based on gender and race attributes, this dissertation proposes and evaluates CL as an effective strategy to ensure fairness in Facial Expression Recognition (FER) systems, guarding against biases arising from imbalances in data distributions. Benchmark comparisons against SOTA ML-based approaches highlight the superior bias-mitigation capabilities of CL-based methods. Finally, this dissertation explores sensing and adapting to human affective behaviour during wellbeing coaching sessions as an application scenario for continually learning affective robots. Enabling personalised human-robot interactions, the Pepper robot is embedded with CLIFER-based facial affect perception, allowing it to personalise its learning towards individual affective behaviour, dynamically adapting the interaction flow to generate naturalistic responses, sensitive to the participants’ affective state. To evaluate such continual personalisation ability in Pepper, a user study is conducted with 20 participants demonstrating that using CL-based personalisation significantly improves the subjective experience of the participants interacting with Pepper. The theoretical formulations, benchmarks, and frameworks presented in this dissertation initiate a novel field of enquiry exploring the benefits of CL-based learning for affective robots. This dissertation aims to create a stepping stone for affective robotics research to consider taking a continual and personalised learning approach towards building fully autonomous and adaptive robots that are purposeful and engaging in their interactions with human users.

Given the challenges associated with the real-world deployment of Machine Learning (ML) models, especially towards efficiently integrating novel information on the go, both Continual Learning (CL) and Causality have been proposed and investigated individually as potent solutions. Despite their complementary nature, the bridge between them is still largely unexplored. In this work, we focus on causality to improve the learning and knowledge preservation capabilities of CL models. In particular, positing Causal Replay for knowledge rehearsal, we discuss how CL-based models can benefit from causal interventions towards improving their ability to replay past knowledge in order to mitigate forgetting., EPSRC, Alan Turing Institute, Cambridge Commonwealth Trust

Understanding human affective behaviour, especially in the dynamics of real-world settings, requires Facial Expression Recognition (FER) models to continuously adapt to individual differences in user expression, contextual attributions, and the environment. Current (deep) Machine Learning (ML)-based FER approaches pre-trained in isolation on benchmark datasets fail to capture the nuances of real-world interactions where data is available only incrementally, acquired by the agent or robot during interactions. New learning comes at the cost of previous knowledge, resulting in catastrophic forgetting. Lifelong or Continual Learning (CL), on the other hand, enables adaptability in agents by being sensitive to changing data distributions, integrating new information without interfering with previously learnt knowledge. Positing CL as an effective learning paradigm for FER, this work presents the Continual Facial Expression Recognition (ConFER) benchmark that evaluates popular CL techniques on FER tasks. It presents a comparative analysis of several CL-based approaches on popular FER datasets such as CK+, RAF-DB, and AffectNet and present strategies for a successful implementation of ConFER for Affective Computing (AC) research. CL techniques, under different learning settings, are shown to achieve state-of-the-art (SOTA) performance across several datasets, thus motivating a discussion on the benefits of applying CL principles towards human behaviour understanding, particularly from facial expressions, as well the challenges entailed.

Sustaining real-world human-robot interactions requires robots to be sensitive to human behavioural idiosyncrasies and adapt their perception and behaviour models to cater to these individual preferences. For affective robots, this entails learning to adapt to individual affective behaviour to offer a personalised interaction experience to each individual. Continual Learning (CL) has been shown to enable real-time adaptation in agents, allowing them to learn with incrementally acquired data while preserving past knowledge. In this work, we present a novel framework for real-world application of CL for modelling personalised human-robot interactions using a CL-based affect perception mechanism. To evaluate the proposed framework, we undertake a proof-of-concept user study with 20 participants interacting with the Pepper robot using three variants of interaction behaviour: static and scripted, using affect-based adaptation without personalisation, and using affect-based adaptation with continual personalisation. Our results demonstrate a clear preference in the participants for CL-based continual personalisation with significant improvements observed in the robot's anthropomorphism, animacy and likeability ratings as well as the interactions being rated significantly higher for warmth and comfort as the robot is rated as significantly better at understanding how the participants feel., Comment: Accepted at the 10th International Conference on Affective Computing and Intelligent Interaction Workshops and Demos (ACIIW). 12 pages, 7 figures

Real-world Facial Expression Recognition (FER) systems require models to constantly learn and adapt with novel data. Traditional Machine Learning (ML) approaches struggle to adapt to such dynamics as models need to be re-trained from scratch with a combination of both old and new data. Replay-based Continual Learning (CL) provides a solution to this problem, either by storing previously seen data samples in memory, sampling and interleaving them with novel data (rehearsal) or by using a generative model to simulate pseudo- samples to replay past knowledge (pseudo-rehearsal). Yet, the high memory footprint of rehearsal and the high computational cost of pseudo-rehearsal limit the real-world application of such methods, especially on resource-constrained devices. To address this, we propose Latent Generative Replay (LGR) for pseudo-rehearsal of low-dimensional latent features to mitigate forgetting in a resource-efficient manner. We adapt popular CL strategies to use LGR instead of generating pseudo-samples, resulting in performance upgrades when evaluated on the CK+, RAF-DB and AffectNet FER benchmarks where LGR significantly reduces the memory and resource consumption of replay-based CL without compromising model performance., EPSRC

Current Humanoid Service Robot (HSR) behaviours mainly rely on static models that cannot adapt dynamically to meet individual customer attitudes and preferences. In this work, we focus on empowering HSRs with adaptive feedback mechanisms driven by either implicit reward, by estimating facial affect, or explicit reward, by incorporating verbal responses of the human ���customer���. To achieve this, we first create a custom dataset, annotated using crowd-sourced labels, to learn appropriate approach (positioning and movement) behaviours for a Robo-waiter. This dataset is used to pre-train a Reinforcement Learning (RL) agent to learn behaviours deemed socially appropriate for the robo-waiter. This model is later extended to include separate implicit and explicit reward mechanisms to allow for interactive learning and adaptation from user social feedback. We present a within-subjects Human-Robot Interaction (HRI) study with 21 participants implementing interactions between the robo-waiter and human customers implementing the above-mentioned model variations. Our results show that both explicit and implicit adaptation mechanisms enabled the adaptive robo-waiter to be rated as more enjoyable and sociable, and its positioning relative to the participants as more appropriate compared to using the pre-trained model or a randomised control implementation., E. McQuillin was supported by the 2020/21 DeepMind Cambridge Scholarship. N. Churamani is funded by the EPSRC grant EP/R513180/1 (ref. 2107412). H. Gunes is supported by the EPSRC project ARoEQ under grant ref. EP/R030782/1.

Collaborative interactions require social robots to share the users’ perspective on the interactions and adapt to the dynamics of their affective behaviour. Yet, current approaches for affective behaviour generation in robots focus on instantaneous perception to generate a one-to-one mapping between observed human expressions and static robot actions. In this paper, we propose a novel framework for affect-driven behaviour generation in social robots. The framework consists of (i) a hybrid neural model for evaluating facial expressions and speech of the users, forming intrinsic affective representations in the robot, (ii) an Affective Core, that employs self-organising neural models to embed behavioural traits like patience and emotional actuation that modulate the robot’s affective appraisal, and (iii) a Reinforcement Learning model that uses the robot’s appraisal to learn interaction behaviour. We investigate the effect of modelling different affective core dispositions on the affective appraisal and use this affective appraisal as the motivation to generate robot behaviours. For evaluation, we conduct a user study (n = 31) where the NICO robot acts as a proposer in the Ultimatum Game. The effect of the robot’s affective core on its negotiation strategy is witnessed by participants, who rank a patient robot with high emotional actuation higher on persistence, while an impatient robot with low emotional actuation is rated higher on its generosity and altruistic behaviour.

Most state-of-the-art approaches for Facial Action Unit (AU) detection rely on evaluating static frames, encoding a snapshot of heightened facial activity. In real-world interactions, however, facial expressions are more subtle and evolve over time requiring AU detection models to learn spatial as well as temporal information. In this work, we focus on both spatial and spatio-temporal features encoding the temporal evolution of facial AU activation. We propose the Action Unit Lifecycle- Aware Capsule Network (AULA-Caps) for AU detection using both frame and sequence-level features. While, at the frame- level, the capsule layers of AULA-Caps learn spatial feature primitives to determine AU activations, at the sequence-level, it learns temporal dependencies between contiguous frames by focusing on relevant spatio-temporal segments in the sequence. The learnt feature capsules are routed together such that the model learns to selectively focus on spatial or spatio-temporal information depending upon the AU lifecycle. The proposed model is evaluated on popular benchmarks, namely BP4D and GFT datasets, obtaining state-of-the-art results for both., EPSRC grant EP/R513180/1 (ref. 2107412). EPSRC project ARoEQ under grant ref. EP/R030782/1. European Union’s Horizon 2020 research and innovation programme WorkingAge project under grant agreement No. 826232.

Current state-of-the-art models for automatic Facial Expression Recognition (FER) are based on very deep neural networks that are effective but rather expensive to train. Given the dynamic conditions of FER, this characteristic hinders such models of been used as a general affect recognition. In this paper, we address this problem by formalizing the FaceChannel, a light-weight neural network that has much fewer parameters than common deep neural networks. We introduce an inhibitory layer that helps to shape the learning of facial features in the last layer of the network and thus improving performance while reducing the number of trainable parameters. To evaluate our model, we perform a series of experiments on different benchmark datasets and demonstrate how the FaceChannel achieves a comparable, if not better, performance to the current state-of-the-art in FER. Our experiments include cross-dataset analysis, to estimate how our model behaves on different affective recognition conditions. We conclude our paper with an analysis of how FaceChannel learns and adapt the learned facial features towards the different datasets., Comment: Accepted for publication at SN Computer Science. arXiv admin note: substantial text overlap with arXiv:2004.08195

Social robots are starting to become incorporated into daily lives by assisting in the promotion of physical and mental wellbeing. This paper investigates the use of social robots for delivering mindfulness sessions. We created a teleoperated robotic platform that enables an experienced human coach to conduct the sessions in a virtual manner by replicating upper body and head pose in real time. The coach is also able to view the world from the robot's perspective and make a conversation with participants by talking and listening through the robot. We studied how participants interacted with a teleoperated robot mindfulness coach over a period of 5 weeks and compared with the interactions another group of participants had with a human coach. The mindfulness sessions delivered by both types of coaching invoked positive responses from the participants for all the sessions. We found that the participants rated the interactions with human coach consistently high in all aspects. However, there is a longitudinal change in the ratings for the interaction with the teleoperated robot for the aspects of motion and conversation. We also found that the participants' personality traits -- conscientiousness and neuroticism influenced the perceptions of the robot coach., Comment: 11 pages. Corrected Typos

Collaborative interactions require social robots to adapt to the dynamics of human affective behaviour. Yet, current approaches for affective behaviour generation in robots focus on instantaneous perception to generate a one-to-one mapping between observed human expressions and static robot actions. In this paper, we propose a novel framework for personality-driven behaviour generation in social robots. The framework consists of (i) a hybrid neural model for evaluating facial expressions and speech, forming intrinsic affective representations in the robot, (ii) an Affective Core, that employs self-organising neural models to embed robot personality traits like patience and emotional actuation, and (iii) a Reinforcement Learning model that uses the robot's affective appraisal to learn interaction behaviour. For evaluation, we conduct a user study (n = 31) where the NICO robot acts as a proposer in the Ultimatum Game. The effect of robot personality on its negotiation strategy is witnessed by participants, who rank a patient robot with high emotional actuation higher on persistence, while an inert and impatient robot higher on its generosity and altruistic behaviour., Comment: 12 pages, 9 figures; An updated version of this article accepted at Frontiers in Robotics and AI as Affect-Driven Learning of Robot Behaviour for Collaborative Human-Robot Interactions

The purpose of the present study is to learn emotion expression representations for artificial agents using reward shaping mechanisms. The approach takes inspiration from the TAMER framework for training a Multilayer Perceptron (MLP) to learn to express different emotions on the iCub robot in a human-robot interaction scenario. The robot uses a combination of a Convolutional Neural Network (CNN) and a Self-Organising Map (SOM) to recognise an emotion and then learns to express the same using the MLP. The objective is to teach a robot to respond adequately to the user's perception of emotions and learn how to express different emotions., Comment: Published in the Workshop on Bio-inspired Social Robot Learning in Home Scenarios, IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Daejeon, Korea (2016)

Most state-of-the-art approaches for Facial Action Unit (AU) detection rely upon evaluating facial expressions from static frames, encoding a snapshot of heightened facial activity. In real-world interactions, however, facial expressions are usually more subtle and evolve in a temporal manner requiring AU detection models to learn spatial as well as temporal information. In this paper, we focus on both spatial and spatio-temporal features encoding the temporal evolution of facial AU activation. For this purpose, we propose the Action Unit Lifecycle-Aware Capsule Network (AULA-Caps) that performs AU detection using both frame and sequence-level features. While at the frame-level the capsule layers of AULA-Caps learn spatial feature primitives to determine AU activations, at the sequence-level, it learns temporal dependencies between contiguous frames by focusing on relevant spatio-temporal segments in the sequence. The learnt feature capsules are routed together such that the model learns to selectively focus more on spatial or spatio-temporal information depending upon the AU lifecycle. The proposed model is evaluated on the commonly used BP4D and GFT benchmark datasets obtaining state-of-the-art results on both the datasets., Comment: Updated Figure 6 and the Acknowledgements. Corrected typos. 11 pages, 6 figures, 3 tables

Being able to recognize emotions in human users is considered a highly desirable trait in Human-Robot Interaction (HRI) scenarios. However, most contemporary approaches rarely attempt to apply recognized emotional features in an active manner to modulate robot decision-making and dialogue for the benefit of the user. In this position paper, we propose a method of incorporating recognized emotions into a Reinforcement Learning (RL) based dialogue management module that adapts its dialogue responses in order to attempt to make cognitive training tasks, like the 2048 Puzzle Game, more enjoyable for the users., Comment: Proceedings of the Workshop on Intelligent Assistive Computing, IEEE World Congress on Computational Intelligence (WCCI) 2018