Machine learning (ML) has recently achieved impressive success in many applications. As ML starts to penetrate into safety-critical domains, security/robustness concerns on ML systems have received lots of attention lately. Very surprisingly, recent work has shown that current ML models are vulnerable to adversarial attacks, e.g. by perturbing the input slightly ML models can be manipulated to output completely unexpected results. Many attack and defence algorithms have been developed in the field under the convenient but questionable Lp attack model.

Machine learning algorithms are being used in a wide range of applications. It is a branch of computer science where the system can learn from the data and make decisions. Financial fraud is an increasing hazard in the financial industry, and it is important to detect a fraudulent transaction. Machine learning algorithms can be used to decide whether the transaction is fraud or not. After the system makes its prediction, it is important for users to understand the reason behind the prediction in such cases.

BlueNode is a SaaS company focused on the sanitation and analysis of marine shipping data. The research project is focused on increasing the precision and accuracy of shipped goods processed through Canadian ports. Should the research prove the be successful, the technical methods used with be directly incorporated into the BlueNode system.

The project investigates how collaborative tasks can be enhanced in AR environments. The intern will develop three approaches to present shared information in a co-located AR setting and conduct usability studies comparing these approaches.

Fraudulent activities are hard to detect, but they cost financial institutions millions of dollars in monetary losses and legal costs every year. Millions of dollars are being lost in credit transactions as criminals are finding new, more sophisticated ways to conduct financial crime. This research project examines novel ways of detecting fraudulent behavior using powerful tools such as Recurrent Neural Networks, a type of machine learning model that is well suited for sequence or historical data.

45 Drives—a Nova Scotia based company—offers a high-density, low-cost data storage solution called the Storinator. While this product has been very successful, clients have indicated they would like a clustered solution which offers similar performance and redundancy, without sacrificing security or drastically increasing the cost. Researchers at the University of New Brunswick have been identified as a good fit for creating a clustered software-architecture in tandem with 45 Drives’ hardware-architecture.

Financial fraud is a serious issue that is taking place globally and causing considerable damage at great expense. Statistical analysis and machine learning tools can help financial institutions detect different types of fraud. In some cases however, mislabeling and the cost of classification may actually increase the volume of ‘false positives’ for supervised methods. As the number of normal transactions in financial domains far outweigh the number of anomalous transactions, it is challenging to classify the anomaly labels.

Internet of things (IoT) includes of multitude of sensors from a wide variety of applications. These sensors produce high volume and high velocity data. Recently there has been much interest in application of such technologies to improve energy management and agricultural practices. The sensors that are installed in the field transmit real time data regarding numerous environmental variables of interest. This data is then used to forecast a future state and to make a well informed business/operation decision according to an expected future state.

Current research in semantic parsing suffers from lack of annotated data, which is hard to acquire. In this project, we aim at tackling the problem of converting natural language utterances to SQL language (Text-to-SQL) on complex databases in a low-resource environment. More specifically, we focus on the research of how multi-task learning (MTL) can help in this task. We will first identify the related natural language processing (NLP) tasks that can contribute to improving the performance of semantic parsing.