We are excited to see 5 of our papers published at the International Conference on Learning Representations (ICLR) 2023. 

ICLR is a top-tier machine learning conference dedicated to the advancement of the branch of artificial intelligence called representation learning, but generally referred to as deep learning. It is well regarded for presenting and publishing cutting-edge research on all aspects of deep learning used in the fields of artificial intelligence, statistics and data science, as well as important application areas such as machine vision, speech recognition, text understanding, gaming, and more.

RBC Borealis at ICLR 2023

RBC Borealis’s published papers in our product and research focus areas at ICLR include novel algorithms for selective nets and time series analysis, key areas for our product work, and other contributions. The list of RBC Borealis’s ICLR papers is below:

Towards Better Selective Classification 📄

Leo Feng, Mohamed Osama Ahmed, Hossein Hajimirsadeghi, Amir Abdi

Latent Bottlenecked Attentive Neural Processes 📄

Leo Feng, Hossein Hajimirsadeghi, Yoshua Bengio, Mohamed Osama Ahmed

Meta Temporal Point Processes 📄

Wonho Bae, Fred Tung, Mohamed Osama Ahmed, Gabriel Oliveira

Scaleformer: Iterative Multi-scale Refining Transformers for Time Series Forecasting 📄

Amin Shabani, Amir Abdi, Lili Meng, Tristan Sylvain

An Equal-Size Hard EM Algorithm for Diverse Dialogue Generation 📄

Yuqiao Wen, Yongchang Hao, Yanshuai Cao, Lili Mou

These projects and types of work are a team effort. At RBC Borealis, we have created a culture of collaboration and a unique combination of factors that help our researchers succeed, including:

  • Partnerships: academic and ecosystem partners: MILA, Amii, UBC, and SFU are all represented in our co-authors.
  • RBC Borealis’s thriving Internship program: Four of the ICLR papers this year were authored by RBC Borealis interns.
  • Resources: Excellent compute infrastructure that allows researchers to run experiments and conduct the necessary tests.

Research at RBC Borealis

Researchers at RBC Borealis conduct research in artificial intelligence for financial services. The research team within RBC Borealis is large team with backgrounds across artificial intelligence including computer vision, machine learning, and natural language processing, with PhDs in computer science, physics, computational finance, mathematics and more.

In 2023, the team’s commitment to creating real-world impact through scientific pursuit led to RBC Borealis establishing research areas of focus – what we call our North Star research – in Asynchronous Temporal Models (ATOM), Non-Cooperative Learning in Competing Markets, and Causal Machine Learning from Observational Data. The latest updates from the ATOM team can be found ATOM: Asynchronous Temporal Models, alongside more information on RBC Borealis’s approach to conducting cutting-edge research.

For all the latest research papers, please explore the full research publications library. 📖