Novel approaches in autonomous learning and simulation of human behaviour - ON-078
Preferred Disciplines: Computer Sciences, Engineering, Post-Doc
Project Length: 4 months
Desired start date: As soon as possible
Location: Toronto, ON
No. of Positions: 1
About the Company:
RBC Research is the research arm of the Royal Bank of Canada. The team’s mandate is to advance the state of the art in financial technologies by conducting research in machine learning and computer vision. RBC is Canada’s largest financial institution with over 80,000 employees across Banking, Insurance, Wealth Management, Investor & Treasury Services, as well as Capital Markets.
Explore machine learning approaches to simulate or approximate a human analyst’s capacity to observe and assimilate knowledge of interdependencies in the world around them through reading, with the aim of building a prototype demonstrating reasoning & planning in reaction to future events utilising acquired knowledge.
Researchers will have access to a large volume of articles, research & analysis from reliable sources used by RBC (approx. 2.5 million items in English per year, with additional volume in other languages, and a 13-year archive) and will be free to select methods, technologies and approaches to experiment with. Researchers may need to consider utilising natural language processing (in particular relationship extraction in text), engineering ontologies and the utilisation of recurrent neural networks.
- Demonstrable practical application of autonomous learning & reasoning
- Build sentiment analysis pipeline for business related news and reports
- Develop deep learning based NLP pipelined to understand text and assign positive, negative or neutral score to the effect the text has on a specific industry
- Build a platform that integrates scores from multiple sources of heterogeneus data to predict the effect of all currently available information on business sector and specific business clients.
- Develop and apply deep learning based techniques (specically recurent networks) for natural language processing (NLP) to business related reports in order to extract useful abstract features from these sources. Apply sentiment analysis to these features to classify whether text sources are positive or negative for specific industries and businesses.
- Develop data integration techniques (potentially using deep learning) to combine multiple heterogenius data sources to predict the future performance and needs of businesses based on currenlty available and previously reported information.
Expertise and Skills Needed:
- Extensive knowledge of artificial intelligence concepts and practical applications, in particular natural language understanding, knowledge representation, reasoning and planning.
- Fuent in Python, R
- Experience with deep learning frameworks (Caffe, Tensorflow, Theano, Torch7)
For more info or to apply to this applied research position, please