A parallel algorithm for quantum circuit synthesis
Quantum circuit synthesis is an important step in the process of quantum compilation. Given an arbitrary unitary operation, quantum circuit synthesis is the process that constructs a quantum circuit using only gates from a universal gate set which is either exactly, or approximately equivalent to the original operation. The currently known algorithms for multi-qubit circuit synthesis run in exponential time and rely on the generation of databases many GBs in size to complete the search; synthesis of circuits of more than 3 qubits up to a certain length is infeasible using current methods due to this exponential scaling. We have developed a framework and accompanying software that uses time/memory tradeoffs and parallel collision finding techniques to synthesize circuits. A simple implementation using 16 OpenMP threads found that this approach can increase the speed of synthesis over previous algorithms, as well as synthesize larger circuits. In order to fully reap the benefits offered by this algorithm, we are developing a hybrid OpenMP/MPI algorithm, with the hopes of scaling up this method to thousands of cores or more.
Applications of Data-Driven Analytics and Decision-Making Optimization for Solar-powered Energy Hubs Integrated with Electrical Vehicles
This project will explore data-based solutions regarding uncertainties and prediction issues associated with the Photovoltaic (PV) systems power output, EV charging load and the buildings’ demand load by utilizing the recent advancements in big data analytics. A data-driven method will be proposed and applied to Boxbrite’s energy systems planning and scheduling framework (e.g., capacity expansion models representing the addition of energy storage batteries, optimal scheduling of building and EV charging loads, minimizing the imported power from the grid etc.). Therefore, high-level optimization tasks such as planning (designing, sizing) and scheduling (operating mode, on-off status) will highly benefit from information mined from massive data, since optimization has always depended on the interchange between models and data. In particular, in this research, a comprehensive framework for microgrids with EV charger systems that can leverage conclusions from big data tools (machine learning) will be developed.
Automatic Identification of Phytoplankton using Deep Learning
Under the impact of global climate changes and human activities, harmful algae blooms (HABs) have become a growing concern due to negative impacts on water related industries, such as safe water supply for fish farmers in aquaculture. The current method of algae identification requires water samples to be sent to a human expert to identify and count all the organisms. Typically this process takes about 1 week, as the water sample must be preserved and then shipped to a lab for analysis. Once at the lab, a human expert must manually look through a microscope, which is both time-consuming and prone to human error. Therefore reliable and cost effective methods of quantifying the type and concentration of algae cells has become critical for ensuring successful water management. Blue Lion Labs, in partnership with the University of Waterloo, is building an innovative system to automatically classify multiple types of algae in-situ and in real-time by using a custom imaging system and deep learning. This will be accomplished using two main steps. First, the technical team will work with an in-house algae expert (a phycologist) to build up a labelled database of images. Second, the research team will design and build a novel neural network architecture to segment and classify the images generated by the imaging system. The result of this proposed research will dramatically reduce the analysis time of a water sample from weeks to hours, and therefore will enable fish farmers to better manage harmful algae blooms.
Cloud‐based data analytic platform for real‐world evidence generation
Randomized controlled trials (RCTs) are considered the gold standard for supportive clinical evidence, however RCTs rarely describe the effectiveness of an intervention in real life practice. As such, regulators, payers, and health‐care providers are turning towards real world data (RWD) to understand how well an intervention performs in clinical practice.
The best source of RWD is source data – that is, data that are collected at the interface of the patient and the health care system, as well patient monitoring and self‐reported data captured directly within the patients’ home environment. Unfortunately, these data are stored in different and distinct systems that are not well integrated and in formats that do not allow for rapid analytic capabilities. The University of Waterloo in partnership with Roche Canada, are therefore proposing to develop the “CARE” (Clinical Analytics for Real‐World Evidence) platform.
The “CARE” platform will be a holistic cloud‐based data analytic solution featuring a large central repository for consolidating data obtained from disparate data systems (including data from electronic medical records, lab data, physician transcriptions, and patient monitoring devices and self‐reported surveys). The “CARE” platform will serve as a central hub for researchers that includes integrated and sophisticated data analytics, access control, security and study management tools in order to curate data for research and clinical purposes. This project begins the initial steps to address relevant research objectives in lung cancer by providing the tool and infrastructure required to process and analyze the enormous amount of scattered oncology data within an institution and across multiple institutions. Ultimately, this work will make it possible to mine currently siloed and/or unstructured data across the system and produce data‐driven insights in order to deliver the right care to the right patient at the right time through scientific innovation and research excellence.
Cold-region Land-surface Simulations under Climate Change
This proposal is part of a larger modelling project, Canada 1 Water, for which separate RAC applications have been submitted. This request is in direct support of a Mitacs application and concerns a subset of simulations that will be performed by or will directly involve the Mitacs interns. It is anticipated that two 150-year climate projection runs using CRCM5 forcing data from the CORDEXarchive, two 50-year historical reanalysis simulations (using ERA5 and NRCan climate forcing) will be necessary, as well as 100 model-years for development, testing and spin-up, totalling 500 model-years.The model that will be used for this project is the Community Terrestrial Systems Model, which is the stand-alone version of CLM5, the land component of the Community Earth System Model v2. One model year of CTSM/CLM5 simulation for Canada at 4km resolution requires approx. 0.2 core-years and 114 GB of storage. This estimate is based on lower-resolution test runs; scaling is trivial since vertical columns do not communicate.
The anticipated 500 model-years would require 100 core-years of compute time and produce 57 TB of output; however, much of the output could also be stored on a separate tape archive or off-site. 50 years of ERA5 climate forcing data would require approximately 20 TB of continuous scratch or project storage; NRCan data and topographic data would require 10 TB, and each 150-year CRCM5 projection would require approximately 16 TB of storage, resulting in 64 TB of forcing data. However, only 50% of that would be required at any given time, so 100 TB of scratch storage would be sufficient. Note that these are estimates and resolutions and the number of simulations can be adjusted as resources permit. Furthermore, the intention is, to make all data publicly available for research.
Distributed and scalable search in enterprise databases
Google search, and other search engines such as Bing and Yahoo!, provide a convenient way to find Webpages that contain various keywords or are related to particular topics. For the purposes of searching, Webpages are essentially loosely structured paragraphs of text. However, much of the world’s high-quality enterprise data are structured into well defined tables containing sets of well-defined columns.
One consequence of structured database design is that information about a single entity may be scattered across many columns in many tables, and must be stitched together in a meaningful way when answering user queries. This turns out to be significantly more difficult than finding Webpages or text documents containing various keywords.
As Dr. Surajit Chadhuri (a Distinguished Scientist at Microsoft Research) recently argued in a keynote talk at the IEEE Data Engineering conference, search over structured databases has fallen behind search over unstructured data. In the proposed research, we will develop a powerful and intuitive search system, akin to Web keyword search, for structured enterprise data. Our system will empower nontechnical users to explore enterprise databases and turn big data into actionable insight, just as Google search has empowered society to explore the Web.
Efficient deep learning for real-time traffic event detection
Miovision is interested in designing the first affordable, low-power, energy efficient real time traffic event detection system that can be installed without the need to be powered by the grid nor the need to be connected directly to city installed infrastructure. Deep learning for traffic event detection can provide overwhelmingly superior accuracy and addresses most of the real-world scenarios that make competing detectors unsuitable for customer adoption. The challenge with deep learning is its complexity, which is currently infeasible for a self-powered real-world embedded detection system. Working with Dr. Alexander Wong and the Vision and Image Processing Lab at the University of Waterloo, the goal of this project is to develop technologies that can significantly reduce the complexity of deep learning for traffic event detection, while maintaining its accuracy and market fit, so that it can be deployed on a low-cost and low-powered hardware platform.
High fidelity simulations and low-order aero-acoustic modeling of engine test cells
The testing and certification of gas turbines demand the well-controlled environment provided by an engine test cell. The resonant acoustic coupling between the flow generated noise from the gas turbine exhaust and the engine test cell impacts the quality and reliability of the engine testing and certification. Predictive modeling of flow generated noise using high-fidelity numerical simulations is central to an a priori acoustic assessment and for the development of noise-mitigating designs. As part of this effort, the Multi-Physics Interaction Lab and University of Waterloo will numerically study, with the help of high fidelity, large-eddy simulations of the SOSCIP high-performance computers, the acoustic noise generation in partially confined jets undergoing a re-acceleration through the test cell ejector system. As a direct outcome, the researchers will develop a low order Aero-acoustic model that will be used by our industrial partner to predict resonant acoustic models to within +/- 20% the frequency and amplitude of the coupling phenomena. This OCE and NSERC-funding project will permit the training and mentoring of four HQP for careers in science and technology within Canada.
Quantum Hybrid quantum-classical simulation and optimization platform for industrials
To address the problems at hand, we are going to start with Theory-Trained Neural networks (TTNs). Shallow theory-trained neural networks have so far been successfully used for learning the solution of highly coupled differential equations in small systems. This project is to scale up both TTNs and Physics-Informed Machine Learning methods for simulating wind dynamics and to build a real-time forecasting platform to optimize wind power generation. Furthermore, we develop even more efficient algorithms by employing theory-trained quantum neural networks, instead of or in addition to the classical neural network. The algorithms we will be employing includes convolutional neural network, Long-Short-Term-Memory network, Variational Quantum Eigensolver, Quantum Monte Carlo, Quantum Approximate Optimization Algorithm. Applying such techniques to large data sets, training deep neural networks, and computational fluid dynamics simulations are computationally heavy and can only be run on HPC in a timely manner.
Hybrid quantum-classical simulation and optimization platform for industrials
Predictable power is valuable power. This is because the intermittent nature of renewables such as wind energy has posed several challenges for planning the daily operation of the electric grid. As their power fluctuates over multiple time scales, the grid operator has to adjust its day-ahead, hour-ahead, and real-time operating procedure. In the absence of cheap and efficient energy storage to make up for sudden power generation shortfalls or excesses on the grid caused by renewables, nowadays, the grid operator sends a signal to power plants approximately every four seconds to ensure a balance between total electric supply and demand. To be able to strengthen the business case for wind power and further drive the adoption of carbon-free energy on electric grids worldwide, novel strategies are required to stabilize the balance of the grid as the number of renewable generators connected to the grid increases.
Our main goal in this collaboration is to design novel models and algorithms that improve an hour or day-ahead renewable energy prediction. As a first step, we are creating a novel weather simulation model using knowledge-guided or Physics Informed Machine Learning (PIML) methods, and machine learning accelerated computational fluid dynamics (CFD) techniques. Furthermore, in parallel, we will be exploring the suitability of current and near-term quantum computers as co-processors to help address the need for high computational power for simulating complex systems such as wind.
Mining microbiome community structure and biomarker identification through data intensive biology, machine learning, and high throughput technologies
Microbial ecosystems such as those associated with the mining industry are complex networks of interacting species and biochemical dependencies. Modeling responses to different or changing environmental conditions is a considerable computational and statistical challenge. Current approaches largely identify important features through differential abundance, ignoring the disparate influence some species or functions exert on the community, greatly simplifying biological complexity. Additionally, these approaches can ignore poorly annotated features (e.g., hypothetical genes, microbial dark matter, ORFans). Excluding these known unknowns and unknown unknowns reduces the resolution and sensitivity of these analyses. Metagenomic feature selection using machine learning has been most widely applied to the human microbiome, which currently has more extensive data than other systems. We will apply a superficially similar but much higher resolution approach to less studied, more dynamic industrial microbiomes, such as mining.
Prototyping and analysis of engines for integrated cryptography and compression
This project will use the Blue Gene/Q platform and the Agile Computing platform to prototype and analyze hardware and software implementations (engines) of novel algorithms for integrated cryptography and compression. Security analysis (cryptanalysis) is computationally very demanding. The high degree of parallelism and computing power offered by the Blue Gene/Q platform will enable the creation and cryptanalysis of powerful new cryptographic algorithms. The Agile Computing platform will provide a platform for prototyping implementation of novel algorithms that combine cryptography and compression on combined CPU+FPGA systems and thereby facilitate collaboration between IBM and the project participants.
Real-time cloud-based hydrologic risk assessment platform development for watershed scale applications
There is a growing recognition within the hydrologic modeling community that the results from real-time hydrologic modeling will need to be analyzed and presented via standardized web/cloud-based tools in a manner that facilitates expeditious interpretation of what could be considered unwieldy large scientific data-sets. Furthermore, there is also growing recognition that best-in-class weather forecast data can potentially add significant value to the resultant hydrologic simulation results.
To support on-going research into real-time hydrologic modeling at Aquanty, the post-doc will develop a pilot application for HGS real-time modeling at a spatial scale relevant to groundwater and surface water management professionals (i.e. >1000 km2) who have interests across agriculture, urban, and industrial water issues. However, there are still technical challenges that must be overcome before real time fully-integrated hydrologic modeling can become operational at a scale large enough to attract significant end user commercial interest. In the project herein, data assimilation methodology will be developed in order to facilitate using high resolution, spatially distributed weather forecast data for multiple time frames (i.e. 1 d, 3d, 7d, 10d), as the principle driver for watershed scale (~4000 km2) fully integrated real-time hydrologic modeling. Data analytics and visualization methodology will also be developed so that large (>1 TB) model output data-sets can be readily interpreted via a cloud hosted dashboard platform. The outcome from the effort will be in the form of a pilot demonstration of a cloud based hydrologic forecasting system.
Water quality analytics, reporting and forecasting using mobile water kit
Real-time monitoring of water quality for bacterial contamination is difficult with present technologies in the market. Assessing water quality for bacterial contamination takes a long time, costly, tiresome, laboratory-based, and mostly not available to where testing and results are needed most. Water samples are tested by municipalities at specified locations and one or twice in a year. This is mainly due to time, cost and complexity involved in testing water samples for bacterial contamination.
The data collected by the regulators is not sufficient to estimate the future trends of water quality. Clearly, there remains a need for a rapid and reliable drinking water quality monitoring for more remote communities and mitigating future illness and outbreak risks in these and other rural or remote communities. Mobile Water Kit (MWK) is a rapid and low-cost test kit that can detect indicator bacteria (E. coli) in water samples within one hour. MWK is a simple method and it will be an optimal solution for testing water samples for bacterial contamination on daily basis.
The proposed project will utilize the functionality of MWK for creating water quality data management for bacterial contamination. We will develop m-Water APP for retrieving the water quality data from MWK and web-console for analyzing the retrieved data over cloud platform. We use the data collected with MWK for analyzing the trends in water quality over a time period. In addition, we will map the water quality data and forecast the E. coli outbreaks with the help of stored data. This project will provide an early warning signal about water quality to the communities, regulators, and/or municipalities. This kind of real-time monitoring of water quality not only empowers the communities with access to clean water but also delivers much-needed intervention for public health.
Wire-free continuous respiratory monitoring using functional bed sheet
Apnea is a temporary stop in breathing and the most common type of apnea is Obstructive Sleep Apnea (OSA), which is a disturbance in sleep that is a leading cause in health problems such as heart disease, diabetes, high blood pressure, and Parkinson’s disease if left unnoticed or untreated. Importantly, 80% of people affected by OSA remain undiagnosed. In infants, a stop in breathing for 20 seconds becomes fatal. Current methods of detection of respiration in hospital settings are either quite invasive, with wires and electrodes attached on the body, or are very time-consuming, such as manual counting method and listening of changes in breathing patterns.
What if abnormal respiratory patterns can be detected non-invasively and provide early prediction of emergencies before they occur? This is the goal of Studio 1 Labs, a health technology company that has developed a responsive bed sheet using a noninvasive method of monitoring respiration patterns and changes in respiratory rate. Capturing millions of data points across a range of sensors, Studio 1 Labs is collaborating with Dr. Plinio Morita from the University of Waterloo, to develop a machine-learning algorithm for detection of respiration changes and to translate big data analysis into useful information, to provide patient-centered care for hospitals and homes.