The emergence of artificial intelligence (AI) driven keyword spotting (KWS) technologies has revolutionized human to machine interaction. Yet, the challenge of end-to-end energy efficiency, memory footprint and system complexity of current neural network (NN) powered AI-KWS pipelines has remained ever present. This paper evaluates KWS utilizing a learning automata powered machine learning algorithm called the Tsetlin Machine (TM). Through significant reduction in parameter requirements and choosing logic over arithmetic-based processing, the TM offers new opportunities for low-power KWS while maintaining high learning efficacy. In this paper, we explore a TM-based keyword spotting (KWS) pipeline to demonstrate low complexity with faster rate of convergence compared to NNs. Further, we investigate the scalability with increasing keywords and explore the potential for enabling low-power on-chip KWS.

Limited availability of labeled data for machine learning on multimodal time-series extensively hampers progress in the field. Self-supervised learning (SSL) is a promising approach to learn data representations without relying on labels. However, existing SSL methods require expensive computations of negative pairs and are typically designed for single modalities, which limits their versatility. We introduce CroSSL (Cross-modal SSL), which puts forward two novel concepts: masking intermediate embeddings produced by modality-specific encoders, and their aggregation into a global embedding through a cross-modal aggregator CroSSL allows for handling missing modalities and end-to-end cross-modal earning without requiring prior data preprocessing for handling missing inputs or negative-pair sampling for contrastive learning. We evaluate our method on a wide range of data, including motion sensors such as accelerometers or gyroscopes and biosignals (heart rate, electroencephalograms, electromyograms, electrooculograms, and electrodermal). Overall, CroSSL outperforms previous SSL and supervised benchmarks using minimal labeled data, and also sheds light on how latent masking can improve cross-modal learning.

Billions of distributed, heterogeneous, and resource constrained IoT devices deploy on-device machine learning (ML) for private, fast, and offline inference on personal data. On-device ML is highly context dependent and sensitive to user, usage, hardware, and environment attributes. This sensitivity and the propensity toward bias in ML makes it important to study bias in on-device settings. Our study is one of the first investigations of bias in this emerging domain and lays important foundations for building fairer on-device ML. We apply a software engineering lens, investigating the propagation of bias through design choices in on-device ML workflows. We first identifyreliability biasas a source of unfairness and propose a measure to quantify it. We then conduct empirical experiments for a keyword spotting task to show how complex and interacting technical design choices amplify and propagatereliability bias. Our results validate that design choices made during model training, like the sample rate and input feature type, and choices made to optimize models, like light-weight architectures, the pruning learning rate, and pruning sparsity, can result in disparate predictive performance across male and female groups. Based on our findings, we suggest low effort strategies for engineers to mitigate bias in on-device ML.

Large-scale mobile data studies can reveal valuable insights into user behavior, which in turn can assist system designers to create better user experiences. After a careful review of existing mobile data literature, we found that there have been no large-scale studies to understand smartphone usage behavior in India -- the second-largest and fastest growing smartphone market in the world. With the goal of understanding various facets of smartphone usage in India, we conducted a mixed-method longitudinal data collection study through an Android app released on Google Play. Our app was installed by 215 users, and logged 11.9 million data points from them over a period of 8 months. We analyzed this rich dataset along the lines of four broad facets of smartphone behavior -- how users use different apps, interact with notihcations, react to different contexts, and charge their smartphones -- to paint a holistic picture of smartphone usage behavior of Indian users. This quantitative analysis was complemented by a survey with 55 users and semi-structured interviews with 26 users to deeply understand their smartphone usage behavior. While our first-of-its-kind study uncovered many interesting facts about Indian smartphone users, we also found striking differences in usage behavior compared to past studies in other geographical contexts. We observed that Indian users spend significant time with their smartphones after midnight, continuously check notifications without attending to them and are extremely conscious about their smartphones’ battery. Perhaps the most dramatic finding is the nature of mobile consumerism of Indian users as shown by our results. Taken together, these and the rest of our findings demonstrate the unique characteristics that are shaping the smartphone usage behavior of Indian users.

We present SensiX++, a multi-tenant runtime for adaptive model execution with integrated MLOps on edge devices, e.g., a camera, a microphone, or IoT sensors. SensiX++ operates on two fundamental principles: highly modular componentisation to externalise data operations with clear abstractions and document-centric manifestation for system-wide orchestration. First, a data coordinator manages the lifecycle of sensors and serves models with correct data through automated transformations. Next, a resource-aware model server executes multiple models in isolation through model abstraction, pipeline automation, and feature sharing. An adaptive scheduler then orchestrates the best-effort executions of multiple models across heterogeneous accelerators, balancing latency and throughput. Finally, microservices with REST APIs serve synthesised model predictions, system statistics, and continuous deployment. Collectively, these components enable SensiX++ to serve multiple models efficiently with fine-grained control on edge devices while minimising data operation redundancy, managing data and device heterogeneity, and reducing resource contention. We benchmark SensiX++ with 10 different vision and acoustics models across various multi-tenant configurations on different edge accelerators (Jetson AGX and Coral TPU) designed for sensory devices. We report on the overall throughput and quantified benefits of various automation components of SensiX++ and demonstrate its efficacy in significantly reducing operational complexity and lowering the effort to deploy, upgrade, reconfigure, and serve embedded models on edge devices.

Breakthroughs in unsupervised domain adaptation (uDA) can help in adapting models from a label-rich source domain to unlabeled target domains. Despite these advancements, there is a lack of research on how uDA algorithms, particularly those based on adversarial learning, can work in distributed settings. In real-world applications, target domains are often distributed across thousands of devices, and existing adversarial uDA algorithms -- which are centralized in nature -- cannot be applied in these settings. To solve this important problem, we introduce FruDA: an end-to-end framework for distributed adversarial uDA. Through a careful analysis of the uDA literature, we identify the design goals for a distributed uDA system and propose two novel algorithms to increase adaptation accuracy and training efficiency of adversarial uDA in distributed settings. Our evaluation of FruDA with five image and speech datasets shows that it can boost target domain accuracy by up to 50% and improve the training efficiency of adversarial uDA by at least 11 times.