We consider the adaptive control problem for discrete-time, nonlinear stochastic systems with linearly parameterised uncertainty. Assuming access to a parameterised family of controllers that can stabilise the system in a bounded set within an informative region of the state space when the parameter is well-chosen, we propose a certainty equivalence learning-based adaptive control strategy, and subsequently derive stability bounds on the closed-loop system that hold for some probabilities. We then show that if the entire state space is informative, and the family of controllers is globally stabilising with appropriately chosen parameters, high probability stability guarantees can be derived.

While reinforcement learning (RL) has shown promising performance, its sample complexity continues to be a substantial hurdle, restricting its broader application across a variety of domains. Imitation learning (IL) utilizes oracles to improve sample efficiency, yet it is often constrained by the quality of the oracles deployed. RPI draws on the strengths of IL, using oracle queries to facilitate exploration--an aspect that is notably challenging in sparse-reward RL-- particularly during the early stages of learning. As learning unfolds, RPI gradually transitions to RL, effectively treating the learned policy as an improved oracle. This algorithm is capable of learning from and improving upon a diverse set of black-box oracles. Integral to RPI are Robust Active Policy Selection (RAPS) and Robust Policy Gradient (RPG), both of which reason over whether to perform state-wise imitation from the oracles or learn from its own value function when the learner's performance surpasses that of the oracles in a specific state. Reinforcement learning (RL) has shown significant advancements, surpassing human capabilities in diverse domains such as Go (Silver et al., 2017), video games (Berner et al., 2019; Mnih et al., 2013), and Poker (Zhao et al., 2022). Despite such achievements, the application of RL is largely constrained by its substantial computational and data requirements and high sample complexity, particularly in fields like robotics (Singh et al., 2022) and healthcare (Han et al., 2023), where the extensive online interaction for trial and error is often impractical. Imitation learning (IL) (Osa et al., 2018) improves sample efficiency by allowing the agent to replace some or all environment interactions with demonstrations provided by an oracle policy.

A major challenge in visually grounded language generation is to build robust benchmark datasets and models that can generalize well in real-world settings. To do this, it is critical to ensure that our evaluation protocols are correct, and benchmarks are reliable. In this work, we set forth to design a set of experiments to understand an important but often ignored problem in visually grounded language generation: given that humans have different utilities and visual attention, how will the sample variance in multi-reference datasets affect the models' performance? Empirically, we study several multi-reference datasets and corresponding vision-and-language tasks. We show that it is of paramount importance to report variance in experiments; that human-generated references could vary drastically in different datasets/tasks, revealing the nature of each task; that metric-wise, CIDEr has shown systematically larger variances than others. Our evaluations on reference-per-instance shed light on the design of reliable datasets in the future.

This paper investigates the implementation of voice-enabled Google Assistant and Amazon Alexa on Raspberry Pi. Virtual Assistants are being a new trend in how we interact or do computations with physical devices. A voice-enabled system essentially means a system that processes voice as an input, decodes, or understands the meaning of that input and generates an appropriate voice output. In this paper, we are developing a smart speaker prototype that has the functionalities of both in the same Raspberry Pi. Users can invoke a virtual assistant by saying the hot words and can leverage the best services of both eco-systems. This paper also explains the complex architecture of Google Assistant and Amazon Alexa and the working of both assistants as well. Later, this system can be used to control the smart home IoT devices.

Model compression has emerged as an important area of research for deploying deep learning models on Internet-of-Things (IoT). However, for extremely memory-constrained scenarios, even the compressed models cannot fit within the memory of a single device and, as a result, must be distributed across multiple devices. This leads to a distributed inference paradigm in which memory and communication costs represent a major bottleneck. Yet, existing model compression techniques are not communication-aware. Therefore, we propose Network of Neural Networks (NoNN), a new distributed IoT learning paradigm that compresses a large pretrained 'teacher' deep network into several disjoint and highly-compressed 'student' modules, without loss of accuracy. Moreover, we propose a network science-based knowledge partitioning algorithm for the teacher model, and then train individual students on the resulting disjoint partitions. Extensive experimentation on five image classification datasets, for user-defined memory/performance budgets, show that NoNN achieves higher accuracy than several baselines and similar accuracy as the teacher model, while using minimal communication among students. Finally, as a case study, we deploy the proposed model for CIFAR-10 dataset on edge devices and demonstrate significant improvements in memory footprint (up to 24x), performance (up to 12x), and energy per node (up to 14x) compared to the large teacher model. We further show that for distributed inference on multiple edge devices, our proposed NoNN model results in up to 33x reduction in total latency w.r.t. a state-of-the-art model compression baseline.

A student learning to speak Mandarin wanders into a marketplace on the streets of China on a sunny summer afternoon. Before long, two vendors approach and begin hawking products, trying to outbid one another. The student must now grasp what's being said and formulate an appropriate response using proper pronunciation to avoid being misunderstood. It's a challenging, yet common, scenario for anyone trying to learn a new language by interacting with native speakers and immersing themselves in a foreign culture. Fortunately, the student in this case is able to pause the unfolding scenario to check the accuracy and tone of her planned response.

Generating comprehensive molecular images of organs and tumors in living organisms can be performed at ultra-fast speed using a new deep learning approach to image reconstruction developed by researchers at Rensselaer Polytechnic Institute. The research team's new technique has the potential to vastly improve the quality and speed of imaging in live subjects and was the focus of an article recently published in Light: Science and Applications, a Nature journal. Compressed sensing-based imaging is a signal processing technique that can be used to create images based on a limited set of point measurements. Recently, a Rensselaer research team proposed a novel instrumental approach to leverage this methodology to acquire comprehensive molecular data sets, as reported in Nature Photonics. While that approach produced more complete images, processing the data and forming an image could take hours.

When a class in Mandarin Chinese starts next summer at Rensselaer Polytechnic Institute, students will be practicing their spoken dialogues with a different sort of teaching assistant: an artificial intelligence chatbot. Capable of conversing with students in simulated settings -- a restaurant, garden or even a Tai Chi class -- the bot is part of a future where artificial intelligence (AI) will perform more of the tasks, and potentially the jobs, now done by humans. Part of a so-called "situations room" at RPI, the chatbot is an example of what are called "cognitive and immersive systems," in which the burgeoning field of AI is melded with rapidly growing torrents of financial, health and education information as well as so-called "unstructured data" like social media posts spreading across an expanding constellation of networked computers, smartphones and other electronic devices. RPI is developing the room under a partnership with the technology giant IBM and its supercomputer Watson, which first gained worldwide attention in 2011 when it beat humans in the TV game show "Jeopardy." It's too early to predict how much impact AI will have on how New Yorkers work, but a recent report by the Albany-based Rockefeller Institute of Government projects that large numbers of jobs being replaced or changed -- particularly in jobs that involve basic, repetitive actions.

IBM Research and Rensselaer Polytechnic Institute (RPI) are collaborating on a new approach to help students learn Mandarin. The strategy pairs an AI-powered assistant with an immersive classroom environment that has not been used previously for language instruction. The classroom, called the Cognitive Immersive Room (CIR), makes students feel as though they are in restaurant in China, a garden, or a Tai Chi class, where they can practice speaking Mandarin with an AI chat agent. The CIR was developed by the Cognitive and Immersive Systems Lab (CISL), a research collaboration between IBM Research and RPI. When learning a new language, especially one as difficult as Mandarin, it's important that students have many opportunities to speak and practice their conversational skills.

IBM Research and Rensselaer Polytechnic Institute (RPI) are collaborating on a new approach to help students learn Mandarin. The strategy pairs an AI-powered assistant with an immersive classroom environment that has not been used previously for language instruction. The classroom, called the Cognitive Immersive Room (CIR), makes students feel as though they are in restaurant in China, a garden, or a Tai Chi class, where they can practice speaking Mandarin with an AI chat agent. The CIR was developed by the Cognitive and Immersive Systems Lab (CISL), a research collaboration between IBM Research and RPI. CISL researchers demonstrate an AI-assisted Mandarin Chinese language learning aid.