What if you could wake up every morning and open the blinds before you crawl out from underneath the covers? Motorized shades are nothing new to the world of home design. But the light-blocking treatments have undergone a convenient--and smart--update in recent years. Like regular window treatments, smart blinds offer privacy, allow you to control the amount of outdoor light coming into your home, and may provide some relief to your energy bill by blocking out heat from the sun. And, just like typical window coverings, smart blinds come in a variety of styles, fabrics, and designs.

The pace of innovation is incredibly fast with new things being discovered daily. This is a special type of intelligence that is exhibited by computers and other machines. It's a flexible agent that perceives its environment and takes the necessary action required for the success of that particular phenomenon. Artificial intelligence is used when machines copy the cognitive functions of the human brain in learning and solving problems. As machines become increasingly capable, other facilities are removed from the definition.

Energy Star was launched in 1992 by the US Environmental Protection Agency as a voluntary labelling programme recognising the value of energy-efficiency in a broad range of computer-related products, from personal computers to air-conditioning systems. The programme's major success was the widespread adoption of the energy-saving "sleep mode" in consumer electronic devices. Energy Star's innovative breakthrough represents an important platform from which today's concept of computational sustainability was launched. Computational sustainability is defined as a field of interdisciplinary research that attempts to optimise societal, economic and environmental resources using advanced decision-making algorithms supported by the ever-increasing processing power of today's evolving computer systems. Computational sustainability's key goals include the development of computational models, methods and tools to assist in the management of the delicate balance between environmental, economic and societal needs. Advancements in AI and HCI have enabled combinations of robots and humans to carry out critical functions in the most hostile of environments.

Deep learning has become an area of interest in most scientific areas, including physical sciences. Modern networks apply real-valued transformations on the data. Particularly, convolutions in convolutional neural networks discard phase information entirely. Many deterministic signals, such as seismic data or electrical signals, contain significant information in the phase of the signal. We explore complex-valued deep convolutional networks to leverage non-linear feature maps. Seismic data commonly has a lowcut filter applied, to attenuate noise from ocean waves and similar long wavelength contributions. Discarding the phase information leads to low-frequency aliasing analogous to the Nyquist-Shannon theorem for high frequencies. In non-stationary data, the phase content can stabilize training and improve the generalizability of neural networks. While it has been shown that phase content can be restored in deep neural networks, we show how including phase information in feature maps improves both training and inference from deterministic physical data. Furthermore, we show that the reduction of parameters in a complex network results in training on a smaller dataset without overfitting, in comparison to a real-valued network with the same performance.

A Semi-Hidden Markov Model (SHMM) for bursty error channels is defined by a state transition probability matrix $A$, a prior probability vector $\Pi$, and the state dependent output symbol error probability matrix $B$. Several processes are utilized for estimating $A$, $\Pi$ and $B$ from a given empirically obtained or simulated error sequence. However, despite placing some restrictions on the underlying Markov model structure, we still have a computationally intensive estimation procedure, especially given a large error sequence containing long burst of identical symbols. Thus, in this paper, we utilize under some moderate assumptions, a Markov model with random state transition matrix $A$ equivalent to a unique Block Diagonal Markov model with state transition matrix $\Lambda$ to model an indoor software-defined power line communication system. A computationally efficient modified Baum-Welch algorithm for estimation of $\Lambda$ given an experimentally obtained error sequence from the indoor PLC channel is utilized. Resulting Equivalent Block Diagonal Markov models assist designers to accelerate and facilitate the procedure of novel PLC systems design and evaluation.

We study the sample complexity of approximate policy iteration (PI) for the Linear Quadratic Regulator (LQR), building on a recent line of work using LQR as a testbed to understand the limits of reinforcement learning (RL) algorithms on continuous control tasks. Our analysis quantifies the tension between policy improvement and policy evaluation, and suggests that policy evaluation is the dominant factor in terms of sample complexity. Specifically, we show that to obtain a controller that is within $\varepsilon$ of the optimal LQR controller, each step of policy evaluation requires at most $(n+d)^3/\varepsilon^2$ samples, where $n$ is the dimension of the state vector and $d$ is the dimension of the input vector. On the other hand, only $\log(1/\varepsilon)$ policy improvement steps suffice, resulting in an overall sample complexity of $(n+d)^3 \varepsilon^{-2} \log(1/\varepsilon)$. We furthermore build on our analysis and construct a simple adaptive procedure based on $\varepsilon$-greedy exploration which relies on approximate PI as a sub-routine and obtains $T^{2/3}$ regret, improving upon a recent result of Abbasi-Yadkori et al.

Mining temporal assertions from time-series data using information theory to filter real properties from incidental ones is a practically significant challenge. The problem is complex for continuous or hybrid systems because the degrees of influence on a consequent from a timed-sequence of predicates (called its prefix sequence), varies continuously over dense time intervals. We propose a parameterized method that uses interval arithmetic for flexibly learning prefix sequences having influence on a defined consequent over various time scales and predicates over system variables.

In the last few years, many asset-intensive organizations, particularly in the mining, power and utilities, oil and gas, and chemicals industries, have turned to industrial Internet of Things (IIoT) and cognitive technologies to help improve a critical area of their business: equipment reliability.1 Asset performance management (APM) programs, which connect data and trigger actions via systems across the business, can play a major part in driving these improvements. According to a 2018 Deloitte survey, oil and gas leaders rated the big data derived from programs such as APM as the most likely to provide the greatest business value.2 However, when asked about how digital technology can be used most effectively within their companies, those same executives ranked APM below both cost reduction in maintenance and operations as well as improvements in safety.3 This seems to reveal a pervasive and narrow view of APM that may miss the connection between asset performance, broader maintenance and operations improvements, and safety. Merely implementing APM software and digitizing existing processes is not likely to improve core operations and obtain the financial results that executive leaders desire (and investors demand).

We consider a new form of model-based reinforcement learning methods that directly learns the optimal control parameters, instead of learning the underlying dynamical system. This includes a form of exploration and exploitation in learning and applying the optimal control parameters over time. This also includes a general framework that manages a collection of such control-model-based reinforcement learning methods running in parallel and that selects the best decision from among these parallel methods with the different methods interactively learning together. We derive theoretical results for the optimal control of linear and nonlinear instances of the new control-model-based reinforcement learning methods. Our empirical results demonstrate and quantify the significant benefits of our approach.

Neuromorphic Computing is a nascent research field in which models and devices are designed to process information by emulating biological neural systems. Thanks to their superior energy efficiency, analog neuromorphic systems are highly promising for embedded, wearable, and implantable systems. However, optimizing neural networks deployed on these systems is challenging. One main challenge is the so-called timescale mismatch: Dynamics of analog circuits tend to be too fast to process real-time sensory inputs. In this thesis, we propose a few working solutions to slow down dynamics of on-chip spiking neural networks. We empirically show that, by harnessing slow dynamics, spiking neural networks on analog neuromorphic systems can gain non-trivial performance boosts on a battery of real-time signal processing tasks.