Large Language Models (LLMs) deployed on edge devices learn through fine-tuning and updating a certain portion of their parameters. Although such learning methods can be optimized to reduce resource utilization, the overall required resources remain a heavy burden on edge devices. Instead, Retrieval-Augmented Generation (RAG), a resource-efficient LLM learning method, can improve the quality of the LLM-generated content without updating model parameters. However, the RAG-based LLM may involve repetitive searches on the profile data in every user-LLM interaction. This search can lead to significant latency along with the accumulation of user data. Conventional efforts to decrease latency result in restricting the size of saved user data, thus reducing the scalability of RAG as user data continuously grows. It remains an open question: how to free RAG from the constraints of latency and scalability on edge devices? In this paper, we propose a novel framework to accelerate RAG via Computing-in-Memory (CiM) architectures. It accelerates matrix multiplications by performing in-situ computation inside the memory while avoiding the expensive data transfer between the computing unit and memory. Our framework, Robust CiM-backed RAG (RoCR), utilizing a novel contrastive learning-based training method and noise-aware training, can enable RAG to efficiently search profile data with CiM. To the best of our knowledge, this is the first work utilizing CiM to accelerate RAG.

One in 36 children in the United States has autism. Numerous robotic intervention systems have been proposed for children with autism. It is widely acknowledged that engagement with the robotic system is important for intervention success. Visual attention toward the robot can be used as a proxy for engagement. However, less is known about how to maintain and enhance visual attention within a robotic system, and how the variation of visual attention may influence adaptation of dynamic intervention protocols to improve outcomes. Therefore, in this work, we propose a new metric, System Capture Ratio (SCR), that can be automatically quantified in real-time, to measure a participant’s visual attention within a robotic system designed for joint attention intervention for children with autism. Then, we demonstrate that compared to a static intervention protocol, a dynamic intervention protocol can help sustain visual attention and thus achieve a significant performance improvement. The results support the implementation of adaptive and dynamic robotic intervention protocols for autistic children that are based on SCR, offering suggestions on designing effective multi-session studies for autism intervention.

We propose a multimodal data fusion framework to systematically analyze human behavioral data from specialized domains that are inherently dynamic, sparse, and heterogeneous. We develop a two-tier architecture of probabilistic mixtures, where the lower tier leverages parametric distributions from the exponential family to extract significant behavioral patterns from each data modality. These patterns are then organized into a dynamic latent state space at the higher tier to fuse patterns from different modalities. In addition, our framework jointly performs pattern discovery and maximum-margin learning for downstream classification tasks by using a group-wise sparse prior that regularizes the coefficients of the maximum-margin classifier. Therefore, the discovered patterns are highly interpretable and discriminative to support downstream classification tasks. Experiments on real-world behavioral data from medical and psychological domains demonstrate that our framework discovers meaningful multimodal behavioral patterns with improved interpretability and prediction performance.

Machine learning-driven human behavior analysis is gaining attention in behavioral/mental healthcare, due to its potential to identify behavioral patterns that cannot be recognized by traditional assessments. Real-life applications, such as digital behavioral biomarker identification, often require the discovery of complex spatiotemporal patterns in multimodal data, which is largely under-explored. To fill this gap, we propose a novel model that integrates uniquely designed Deep Temporal Sets (DTS) with Evidential Reinforced Attentions (ERA). DTS captures complex temporal relationships in the input and generates a set-based representation, while ERA captures the policy network’s uncertainty and conducts evidence-aware exploration to locate attentive regions in behavioral data. Using child-computer interaction data as a testing platform, we demonstrate the effectiveness of DTS-ERA in differentiating children with Autism Spectrum Disorder and typically developing children based on sequential multimodal visual and touch behaviors. Comparisons with baseline methods show that our model achieves superior performance and has the potential to provide objective, quantitative, and precise analysis of complex human behaviors.

OBJECTIVE: This article reports results from a systematic literature review of the current state of mobile health (mHealth) technologies that have the potential to support self-management for people with diabetes and hypertension. The review aims to (a) characterize mHealth technologies used or described in the mHealth literature and (b) summarize their effects on self-management for people with diabetes and hypertension from the clinical and technical standpoints. MATERIALS AND METHODS: A systematic literature review was conducted following PRISMA guidelines. Online databases were searched in September 2018 to identify eligible studies for review that had been published since 2007, the start of the smartphone era. Data were extracted from included studies based on the PICOS framework. RESULTS: Of the 11 studies included for in-depth review, 5 were clinical research examining patient health outcomes and 6 were technology-focused studies examining users' experiences with mHealth technologies under development. The most frequently used mHealth technology features involved self-management support (n = 11) followed by decision support (n = 6) and shared decision-making (n = 6). Most clinical studies reported benefits associated with mHealth interventions. These included reported improvements in objectively measured patient health outcomes (n = 3) and perceptual or behavioral outcomes (n = 4). DISCUSSION: Although most studies reported promising results in terms of the effects of mHealth interventions on patient health outcomes and experience, the strength of evidence was limited by the study designs. CONCLUSION: More randomized clinical trials are needed to examine the promise and limitations of mHealth technologies as assistive tools to facilitate the self-management of highly prevalent comorbidity of chronic conditions, such as diabetes and hypertension.

Sensory abnormalities are prevalent among children with Autism Spectrum Disorder (ASD). Precise and objective assessment of sensory processing is important for ASD diagnosis and intervention. Traditionally, the assessment is done through psychological interviews, observations, and questionnaires. While these methods have important diagnostic utility, they are subject to biases and have limited resolution. Therefore, we explored the potential of using VR, which provided objective measurements and a higher observation resolution, to assess the sensory processing differences between children with ASD and typically developing (TD). For this pilot investigation, we focused on visual and touch processing. The system can record real-time behavioral data in high temporal and spatial resolutions. An experimental study was conducted with six children with ASD and six TD children aged 11-17 years old. Results showed that the system could reveal sensory differences between these two groups. Strong and moderate correlations were observed between the system's detection and that of the most commonly used standardized psychological assessment questionnaire of sensory processing.

Autism Spectrum Disorders (ASD) are characterized by deficits in social communication skills, such as response to joint attention (RJA). Robotic systems have been designed and applied to help children with ASD improve their RJA skills. One of the most important goals of robot-assisted intervention is helping children generalize social interaction skills to interact with other people. Thus predicting children's human-human interaction (HHI) performance based on their human-robot interaction (HRI) process is an important task. However, to the best of our knowledge, little research exists exploring this topic. The Early Social-Communication Scales (ESCS) test is a measurement of nonverbal social skills, including RJA, for young children. We conducted two longitudinal user studies with a robot-mediated RJA system in young children with ASD, followed by HHI sessions consisting of ESCS administration. In this paper, we present findings regarding how to predict participants' RJA performance in HHI based on their head pose patterns in HRI, under a semi-supervised machine learning framework. As a three-class classification problem, we achieved a micro-averaged accuracy of 73.5%, which indicates the potential effectiveness of the proposed method.

Research indicates that human-robot interaction can help children with Autism Spectrum Disorder (ASD). While most early robot-mediated interaction studies were based on free interactions, recent studies have shown that robot-mediated interventions that focus on the core impairments of ASD such as joint attention deficit tend to produce better outcomes. Joint attention impairment is one of the core deficits in ASD that has an important impact in the neuropsychological development of these children. In this work, we propose a novel joint attention intervention system for children with ASD that overcomes several existing limitations in this domain such as the need to use body-worn sensors, non-autonomous robot operation requiring human involvement and lack of a formal model for robot-mediated joint attention interaction. We present a fully autonomous robotic system, called NORRIS, that can infer attention through a distributed non-contact gaze inference mechanism with an embedded Least-to-Most (LTM) robot-mediated interaction model to address the current limitations. The system was tested in a multi-session user study with 14 young children with ASD. The results showed that participants' joint attention skills improved significantly, their interest in the robot remained consistent throughout the sessions, and the LTM interaction model was effective in promoting the children's performance.

Physical embodiment of robots provides users with a social environment. To design social robots further to be accepted as our companions, we need to understand the essential factors and implement them and so, users get to bring them to their personal environments. To this aim, we focused on two important factors in robotic companionship: robot appearance (look) and emotional expression (feel). Twenty-one participants played an online game with the help from two humanoid robots, Nao (more human-like looking) and Darwin (less human-like looking). Participants interacted with each robot either with emotional words or without emotional words. Results show that only when the robot both looks more human-like and speaks with emotional expression, participants perceive it as their companion. Implications are discussed with future works.

Autism Spectrum Disorders (ASD) impact 1 in 88 children in the United States. The cost of ASD intervention is tremendous with huge individual and social consequences. In recent years, robotic systems have been introduced with considerable success for ASD intervention because of their potential to engage children with ASD. In this work, we present a novel closed-loop autonomous robotic system for imitation skill learning for ASD intervention. Children with ASD show powerful impairment in imitation, which has been associated with a host of neurodevelopmental and learning challenges over time. The presented robotic system offers dynamic, adaptive and autonomous interaction for learning of imitation skills with real-time performance evaluation and feedback. The system has been tested in a user study with young children with ASD and typically developing (TD) control sample. Further, the performance of the system was compared with that of a human therapist in the user study. The results demonstrate that the developed robotic system is well-tolerated by the target population, engaged the children with ASD more than a human therapist, and produced performances that were relatively better than that of a human therapist.

Autism Spectrum Disorders (ASD) are characterized by atypical patterns of behaviors and impairments in social communication. Among the fundamental social impairments in the ASD population are challenges in appropriately recognizing and responding to facial expressions. Traditional intervention approaches often require intensive support and well-trained therapists to address core deficits, with many with ASD having tremendous difficulty accessing such care due to lack of available trained therapists as well as intervention costs. As a result, emerging technology such as virtual reality (VR) has the potential to offer useful technology-enabled intervention systems. In this paper, an innovative VR-based facial emotional expression presentation system was developed that allows monitoring of eye gaze and physiological signals related to emotion identification to explore new efficient therapeutic paradigms. A usability study of this new system involving ten adolescents with ASD and ten typically developing adolescents as a control group was performed. The eye tracking and physiological data were analyzed to determine intragroup and intergroup variations of gaze and physiological patterns. Performance data, eye tracking indices and physiological features indicated that there were differences in the way adolescents with ASD process and recognize emotional faces compared to their typically developing peers. These results will be used in the future for an online adaptive VR-based multimodal social interaction system to improve emotion recognition abilities of individuals with ASD.

In this paper, we propose a novel unsupervised evolutionary clustering algorithm for mixed type data, evolutionary k-prototype algorithm (EKP). As a partitional clustering algorithm, k-prototype (KP) algorithm is a well-known one for mixed type data. However, it is sensitive to initialization and converges to local optimum easily. Global searching ability is one of the most important advantages of evolutionary algorithm (EA), so an EA framework is introduced to help KP overcome its flaws. In this study, KP is applied as a local search strategy, and runs under the control of the EA framework. Experiments on synthetic and real-life datasets show that EKP is more robust and generates much better results than KP for mixed type data.