$247 – $547

Toronto Machine Learning Society (TMLS) : 2019 Annual Conference & Expo

Event Information

Share this event

Date and Time



The Carlu

444 Yonge Street

Toronto, ON M5B 2H4


View Map

Refund Policy

Refund Policy

Refunds up to 30 days before event

Event description


TMLS consists of a community comprised of over 6,000 ML researchers, professionals and entrepreneurs.

We'd like to welcome you to join us in celebrating the top achievements in AI Research, AI companies, and applications in industry.

Expect 1 day of workshops and 2-days of high-quality networking, food, drinks, workshops, breakouts, keynotes and exhibitors.

Taken from the real-life experiences of our community, the Steering Committee has selected the top applications, achievements and knowledge-areas to highlight across 2 days, and 2 nights.

Come expand your network with machine learning experts and further your own personal & professional development in this exciting and rewarding field.

Included will be:

  • 2,000+ (total) attendees

  • NEW! App-supported pre-event networking.

  • Hands-on Workshops (Full bonus day)

  • Breakouts and Keynotes on-site

  • 70+ Speakers

  • AI Career Fair with top 50 AI Start-ups and 1200+ job-seekers.

  • Women in Data Science Ceremony

  • Poster Session Awards

  • Social pub networking afterparty

We believe these events should be as accessible as possible and set our ticket passes accordingly

*Please Scroll down for full Program/Abstracts*

The TMLS initiative is dedicated to helping promote the development of AI/ML effectively, and responsibly across all Industries. As well, to help data practitioners, researchers and students fast-track their learning process and develop rewarding careers in the field of ML and AI.

Current Speakers:

Our Sponsors

*See abstracts Below*

*See abstracts Below*

Day 1 Select Abstracts

ML in Production- Applied Case Study Talk: Building an AI Engine for Time Series Data Analytics
Jian Chang, Senior Algorithm Expert, Alibaba Group

  • Abstract: Hundreds of petabytes of time series data are generated each day in many enterprises. It’s challenging to query this rapidly growing data in a timely manner. TSDB is short for "time series database", which can be used the backbone service for hosting all this data to enable high-concurrency storage and low-latency query. An AI engine on TSDB provides intelligent advanced analysis capabilities and end-to-end business intelligence solutions and empowers companies across various industries to better understand data trends, discover anomalies, manage risks, and boost efficiency. We the design of the AI engine to enable fast and complex analytics of large-scale time series data in many business domains. Along the way, they highlight solutions to the major technical challenges in data storage, processing, feature engineering, and machine learning algorithm design.

Advanced Research Talk: Lookahead Optimizer: k steps forward, 1 step back
Michael Zhang Researcher, University of Toronto & Vector Institute

  • Abstract: The vast majority of successful deep neural networks are trained using variants of stochastic gradient descent (SGD) algorithms. Recent attempts to improve SGD can be broadly categorized into two approaches: (1) adaptive learning rate schemes, such as AdaGrad and Adam, and (2) accelerated schemes, such as heavy-ball and Nesterov momentum. In this paper, we propose a new optimization algorithm, Lookahead, that is orthogonal to these previous approaches and iteratively updates two sets of weights. Intuitively, the algorithm chooses a search direction by looking ahead at the sequence of "fast weights" generated by another optimizer. I will discuss how neural network algorithms can be analyzed and show that Lookahead improves the learning stability and lowers the variance of its inner optimizer with negligible computation and memory cost. I will then present empirical results demonstrating Lookahead can significantly improve the performance of SGD and Adam, even with their default hyperparameter settings on ImageNet, CIFAR-10/100, neural machine translation, and Penn Treebank.

Business Talk: Harnessing Graph-native Algorithms to Enhance Machine Learning: A Primer
Brandy Freitas Senior Data Scientist, Pitney Bowes

  • Abstract: Graph databases have become much more widely popularized in the recent year. Brandy Freitas demystifies the mathematical principles behind graph databases, offers a primer to graph native algorithms, and outlines the current use of graph technology in industry.
    By representing highly connected data in a graph, you have access to a host of graph native algorithms that depend on and exploit the relationships between your data. Computing and storing graph metrics can add strong new features to nodes, creating innovative predictors for machine learning. Using algorithms designed for path finding, centrality, community detection, and graph pattern matching, you can rely less on inflexible, subject-driven feature engineering.

  • Beyond use of derived graph metrics, finding a way to incorporate information about the structure of the graph is a critical issue for furthering the use of machine learning on connected data. So, the question is how to enable the machine learning algorithm to access the inherent structure of the graph itself. Similar to the movement in natural language processing (Word2Vec), where the aim is to preserve information about where a word is in a sequence, there’s a movement in graph analysis to capture community and adjacency of nodes. Using node embedding to create a low dimension vector representation of the node and its structural components, you no longer need to compromise and query away important structural relationships.

  • What you’ll learn: An understanding of the uses of native graph algorithms, advantages to using graph derived metrics in feature engineering, and current techniques for encoding graph structural information into low dimensional feature vectors. Also, how graph metrics can provide enhanced features for machine learning, and where graph database technology is appropriate (and where it isn't) in industry use cases

Applied Case Study Talk: Applications of AI in medicine: roadblocks and opportunities
Niki Athanasiadou, Data scientist, H2O.ai

  • Abstract: As data processing and storage is becoming cheaper, the main barrier to entry for AI adoption is often data availability. This couldn’t be better exemplified than in medicine, where advancements in AI-enabled clinical decision support are mirroring innovations of how data are recorded and stored within healthcare systems. AI-enabled clinical decision support includes diagnosis and prognosis, and involves classification or regression algorithms that can predict the probability of a medical outcome or the risk for a certain disease. Several image classification algorithms using medical images have been approved by the FDA as diagnostic tools in the last two years, and more are certain to follow. Similarly, FDA approval has already been given to wearable devices that monitor vital signs to capture irregularities. These early examples demonstrate the huge potential of AI applications in medicine, as the volume and variety of medical data that get captured increases.

    More than 80-90% of US hospitals and physician offices are implementing some form of an EHR, and similar or even higher adoption rates are seen globally. Despite persistent outstanding issues, the lack of interoperability between EHR systems or patient history continuity, past barriers to adoption relating to data usability and availability are being overcome. Three examples of clinical decision support AI models built on EHR data will be discussed. (1) Accumulation of medical histories from birth alongside linked maternal EHR information in a healthcare facility, enabled the prediction of high obesity risk children as early as two years after birth, possibly allowing life-altering preventative interventions. (2) The Advanced Alert Monitoring system developed and deployed by Kaiser Permanente uses Intensive Care Unit (ICU) data to predict fatally deteriorating cases and alert staff to the need of life-saving interventions. (3) Last, but not least, clinical decision support systems are often required to provide sufficient explanations of their predictions. Global and local explanations of predictions regarding hospital readmissions demonstrate how interpretability techniques enable such explanations. As EHR information becomes standardized and enriched with eg. genomic information, medicine is poised to leverage AI breakthroughs to improve health outcomes.

  • What you'll learn: The talk will present published and unpublished use cases of AI in medicine, showcasing opportunities for adoption, as past obstacles are being resolved.

Advanced Research Talk: Differential Equations for Irregularly-Sampled Time Series Differential Equations
David Duvenaud, Assistant Professor University of Toronto, Vector Institute

  • Abstract: Much real-world data is sampled at irregular intervals, but most time series models require regularly-sampled data. Continuous-time latent variables models can handle address this problem, but until now only deterministic models, such as latent ODEs, were efficiently trainable by backprop. We generalize the adjoint sensitivities method to SDEs, constructing an SDE that runs backwards in time and computes all necessary gradients, along with a general algorithm that allows SDEs to be trained by backpropgation with constant memory cost. We also give an efficient algorithm for gradient-based stochastic variational inference in function space, all with the use of adaptive black-box SDE solvers. Finally, we'll show initial results of applying latent SDEs to time series data, and discuss prototypes of infinitely-deep Bayesian neural networks.

Business Talk: Trustworthy AI: Model Validation at Scale
Layli Goldoozian, Data Scientist, Lucy Liu, Director, Greg Kirczenow, Senior Director, Enterprise Model Risk Management, RBC

  • Abstract: Is AI trustworthy? It is with the right validation tools. As AI continues to transform our capabilities, we are increasingly focused on validating that our models are fair and ethical. One of the challenges seen with this new field is that there are several methods and definitions for making AI trustworthy. RBC has worked on developing model validation at scale through self-serve capabilities based on the most updated and trusted research methods. It allows one to reduce validation time, increase model trust, and establish guarantees on specific properties of models. In this presentation, the speakers will talk about how to develop ethical model validation, and integrating model validation and testing into the data science pipeline. They will introduce the key packages including fairness, model performance, model interpretation, auto benchmark building and data validation.

Applied ML Talk: An Explanation of What, Why, and how of Explainable AI (XAI)
Bahador Khaleghi, Customer Data Scientist and Solution Engineer, H2O.ai

  • Abstract: Modern AI systems are increasingly capable of tackling real-world problems. Yet the black box nature of some AI systems, giving results without a reason, is hindering the mass adoption of AI. According to an annual survey by PwC, the vast majority (82%) of CEOs agree that for AI-based decisions to be trusted, they must be explainable. As AI becomes an ever more integral part of our modern world, we need to understand why and how it makes predictions and decisions.

    These questions of why and how are the subject of the field of Explainable AI, or XAI. Like AI itself, XAI isn’t a new domain of research, and recent advances in the theory and applications of AI have put new urgency behind efforts to explain it. In this talk we will present a technical overview of XAI. The presentation will cover the there key questions of XAI: “What is it?”, “Why is it important?”, and “How can it be achieved?”.

    The what of XAI part takes a deep dive into what it really means to explain AI models in terms of existing definitions, the importance of explanation users’ roles and given application, possible tradeoffs, and explanation studies beyond the AI community. In the why of XAI part, we explore some of the most important drivers of XAI research such as establishing trust, regulatory compliance, detecting bias, AI model generalization and debug.
    Finally, in the how of XAI part we discuss how explainability principles can be applied before, during, and after the modelling stage of AI solution development. In particular, we introduce a novel taxonomy of post-modelling explainability methods, which we then leverage to explore the vast XAI literature work.

  • Who is this presentation for: The first two parts of the talk (the What and Why of XAI) are targeted at a broader audience who are not AI experts but are somewhat familiar with AI.
The last part (the How of XAI) part is intended for AI experts and practitioners who would like to learn about applying XAI in their work.

  • What you’ll learn: The content presented in this talk is unique in two ways. The breadth of XAI literature that is covered is quite vast and, yet, it is highly structured to make the material easier to digest. More importantly, it is intended to provide insights to audience with both technical and business background.

Advanced Technical Talk: HoloClean: A Scalable Prediction Engine for Automating Structured Data Prep
Ihab Ilyas Founder, Professor, Tamr, University of Waterloo

  • Abstract: Data scientists spend big chunk of their time preparing, cleaning, and transforming raw data before getting the chance to feed this data to their well-crafted models. Despite the efforts to build robust predication and classification models, data errors still the main reason for having low quality results. This massive labor-intensive exercises to clean data remain the main impediment to automatic end-to-end AI pipeline for data science.

    In this talk, the speaker will focus on data prep and cleaning as an inference problem, which can be automated by leveraging modern abstractions in ML. The speaker will describe the HoloClean framework, a scalable prediction engine for structured data. The framework has multiple successful proof of concepts with cleaning census data, market research data, and insurance records. The pilots with multiple commercial enterprises showed a significant boost to the quality of source (training) data before feeding them to downstream analytics.

    HoloClean builds two main probabilistic models: a data generation model (describing how data was intended to look like); and a realization model (describing how errors might be introduced to the intended clean data). The framework uses few-shot learning, data augmentation, and self supervision to learn the parameters of these models, and use them to predict both error and their possible repairs.

Business Talk: Lessons from Google's Journey to AI-First
Chanchal Chatterjee, Leader in Artificial Intelligence Solutions, Google

  • Speaker Description: Chanchal Chatterjee, Ph.D, held several leadership roles in machine learning, deep learning and real-time analytics. He is currently leading Machine Learning and Artificial Intelligence at Google Cloud Platform. He won 16 awards at Google in last 18 months. Winner of Outstanding Paper Award by IEEE Neural Network Council. Speaker at many international conferences. Author of a book (in progress) on machine learning with UCLA. Author of 29 Patents and 17 Journal publications and 15 Conference publications. In last 18 months completed over 40 customer interactions and successful machine learning customer projects.

Applied ML in Production Use Case: Scaling Machine Learning - Choosing the Right Approach
Razvan Peteanu Lead Architect, Machine Learning, TD Securities

  • Abstract: Some of the libraries very commonly taught and used in data science have not been designed for large scale machine learning so scaling up computation can be a challenge, particularly that many courses tend to focus on the algorithms and do not cover ML engineering. On the positive side, there are many ways to address this today and choosing the right one for a given project is an important decision as changing architectures can be expensive.
    The talk will go through the pros and cons of several approaches to scale up machine learning, including very recent developments.
  • What you’ll learn: The audience will learn what criteria to use to choose the appropriate approach for their case, as well as the practical pros & cons of each.

  • What languages will be discussed: Python and Java. The speaker will discuss multiple infrastructure options

Advanced Technical Talk: Image Augmentations for Semantic Segmentation and Object Detection
Vladimir Iglovikov, Senior Computer Vision Engineer, Lyft

  • Abstract: In his talk, the speaker will cover image augmentations that are used for object detection and semantic segmentation tasks. They will also talk about novel types of transforms that allow achieving state of the art results in research and in deep learning competitions. They will also discuss their applications for different domains such as self-driving, satellite and medical imagery.

Business Panel: Determining Which ML Opportunities You Should Prioritize
Tomi Poutanen, Chief AI Officer, TD, Founder, Layer 6 AI, Simona Gandrabur, Sr. Director, AI Lead at the National Bank of Canada, Wealth Division- Ofer Shai Chief AI Officer Deloitte, Omnia AI, Rupinder Dhillon- Chief Data Officer, SVP Data & AI, Hudson's Bay Company. Moderator: Trishala Pillai, Applied AI Partner Myplanet

  • Description: Choosing which AI projects to invest in can be a daunting task. How do Layer6’s capabilities fit within TD bank? How does ML research get rolled into products at National Bank of Canada? How is Hudson’s Bay Company selecting the AI projects to build out? How has advanced research from Toronto’s Machine Learning Group help clients at Deloitte?

Advanced Research Track: Adversarial Examples and Understanding Neural Network Representation Space
Nick Frosst , rSWE, GoogleBrain

ML in Production - Implementation, Tooling & Engineering, Data/ML Ops: Building Private Machine Learning Models with TensorFlow
Chang Liu, Applied Research Scientist, Georgian Partners

  • Abstract: This talk will introduce differential privacy and its use cases, discuss the new component of the TensorFlow Privacy library, and offer real-world scenarios for how to apply the tools.

    In recent years, the world has become increasingly data-driven and individuals and organizations have developed a stronger awareness and concern for the privacy of their sensitive data. It has been shown that it is impossible to disclose statistical results about a private database without revealing some information. In fact, the entire database could be recovered from a few query results. Following research on the privacy of sensitive databases, a number of big players such as Google, Apple, and Uber have turned to differential privacy to help guarantee the privacy of sensitive data. That attention from major technology firms has helped bring differential privacy out of research labs and into the realm of software engineering and product development. Differential privacy is now something that smaller firms and software startups are adopting and finding great value in. Apart from privacy guarantees, advances in differential privacy also allow businesses to unlock more capabilities and increased data utility. One of these capabilities includes the ability to transfer knowledge from existing data through differentially private ensemble models without data privacy concerns. As differential privacy garners recognition in large tech companies, efforts to make current state-of-the-art research more accessible to the general public and small startups are underway. As a contribution to the broader community, Georgian Partners has provided its differential privacy library to the TensorFlow community. Together, we will make differentially private stochastic gradient descent available in a user-friendly and easy-to-use API that allows users to train private logistic regression

    What you’ll learn: This talk will introduce differential privacy and its use cases, discuss the new component of the TensorFlow Privacy library, and offer real-world scenarios for how to apply the tools.

Advanced Research Track: Temporal Concept Localization on YouTube 8M Dataset
Satya Krishna Gorti Machine Learning Scientist, Layer6 AI

  • What will you learn: Concepts in videos are high level labels given to segments within videos. They can be actions such as "skateboarding", "dancing" or more general entities such as "acoustic guitar", "wedding". This presentation will be an overview of concept recognition and localization task with respect to YouTube-8M dataset. It will cover methods in recent literature and our approach to reach state-of-the-art results on the dataset.

Day 2 Select Abstracts

Business Talk - Transformation from Research Lab to Product Centers
Daniel Weimer, Head of AI Volkswagen of America, Inc

  • Abstract: Learn about the speaker's work VW's AI Product center. How does their Head of AI lead the transformation from lab to product center? How do they approach this transformation? Which tech stack do they use and how do they develop and deploy? Where is AutoML being used?

  • Speaker Bio: Daniel has built and is leading the deep machine learning competence team. He had defined and implementing HW/SW strategy to deliver scalable machine learning products at VW. He has experience deploying machine learning solutions along the whole automotive value chain

ML Case Study Talk - Machine Learning for Space Exploration
Shreyansh Daftry AI Research Scientist, NASA

Advanced Research Talk - Machine Learning for Systems
Azalia Mirhoseini, Senior Research Scientist, Google Brain

  • Abstract: In this talk, The speaker will present some of their recent work at the intersection of machine learning and systems. First, they discuss their work on the sparsely gated mixture of experts, a new conditional neural network architecture that allows us to train models with 130B+ parameters (10x larger than any previous model) on datasets with 100B+ examples. This architecture uses an intelligent gating mechanism that routes input examples to a subset of the modules (“experts”) within the larger model. Even with a moderate number of parameters, this model runs 2-3x faster than top-performing baselines and sets a new state of the art in machine translation and language modeling. Next, they discuss their work on deep reinforcement learning models that learn to do resource allocation, a combinatorial optimization problem that repeatedly appears in computer systems design and operation. Their method is end-to-end and abstracts away the complexity of the underlying optimization space; the RL agent learns the implicit tradeoffs between computation and communication of the underlying resources and optimizes the allocation using only the true reward function (e.g., the runtime of the generated allocation). The complexity of their search space is on the order of 9^80000, compared to 10^360 states for Go (solved by AlphaGo). Finally, they discuss their work on deep models that learn to find solutions for the classic problem of balanced graph partitioning with minimum edge cuts. They define an unsupervised loss function and use neural graph representations to adaptively learn partitions based on the graph topology. Their method enables generalization; they can train models that produce performant partitions at inference time on new unseen graphs. The generalization significantly speeds up the partitioning process over all existing baselines, where the problem is solved from scratch for each new graph.

Business Panel – Autonomous Vehicles and the Future of Mobility
Ted Graham Head of Open Innovation, GM, Shreyansh Daftry AI Research Scientist, NASA, Steven Lake Waslander , Associate Prof, University of Toronto Moderator: Arif Virani, COO, DarwinAI

Applied Machine Learning Case Study TalkData Science at the New York Times,
Christopher Wiggins, Chief Data Scientist, New York Times

  • Abstract: The Data Science group at The New York Times develops and deploys machine learning solutions to newsroom and business problems. Re-framing real-world questions as machine learning tasks require not only adapting and extending models and algorithms to new or special cases but also sufficient breadth to know the right method for the right challenge. The speaker will first outline how unsupervised, supervised, and reinforcement learning methods are increasingly used in human applications for description, prediction, and prescription, respectively. The speaker will then focus on the 'prescriptive' cases, showing how methods from the reinforcement learning and causal inference literatures can be of direct impact in engineering, business, and decision-making more generally.

ML in Production: Implementation, Tooling & Engineering, Data/ML Ops Talk: DevOps for Machine Learning and other Half-Truths: Processes and Tools for the ML Life Cycle
Kenny Daniel, Founder, Algorithmia

  • Abstract: Traditional software development has a roadmap—the Software Development Life Cycle, coalesced around a specific set of tools and processes. In contrast, machine learning development is a tangle of tools, languages, and infrastructures, with almost no standardization at any point in the process. Manual stopgaps and one-off integrations get models into production, but introduce fragility and risk that prevents businesses from trusting them with mission-critical applications. To build and deploy enterprise-ready models that generate real value, businesses need to standardize on a new stack and a new, ML-focused life cycle.

  • This talk will cover:
    - Key differences between ML and traditional software development
    - Where the SDLC works with ML, and where it breaks down
    - An overview of the new ML stack, from training to deployment to production
    - The five biggest infrastructure and process mistakes ML teams commit
    - How successful early movers have succeeded, and lessons you can use today

  • Infastructures discussed: Docker, Kubernetes.
    Languages discussed: Python, R
    DevOps Tools Discussed: Jenkins, Github, potentially others

Advanced Research Talk - Explaining with Impact: A Machine-centric Strategy to Quantify the Performance of Explainability Algorithms
Sheldon Fernandez, CEO at DarwinAI

  • Abstract: The prevailing progress around AI has created in its wake a heightened interest in Explainable Artificial Intelligence (XAI), whose goal is to produce interpretable decisions made by machine learning algorithms. Of particular interest is the interpretation of how deep neural networks make decisions, given the complexity and 'black box' nature of such networks.

    Given the infancy in the field, there has been limited exploration into the assessment of the performance of explainability methods, with most evaluations centered on subjective and visual interpretations of current approaches. In this talk, the speakers introduce two quantitative performance metrics for quantifying the performance of explainability methods on deep neural networks via a novel decision-making impact analysis: 1.) Impact Score, which assesses the percentage of critical factors with either strong confidence reduction impact or decision changing impact; and 2.) Impact Coverage, which assesses the percentage coverage of adversarially impacted factors in the input. We further consider a comprehensive analysis using this approach against numerous state-of-the-art explainability methods.

Applied ML in Production Case Study Talk- Rearchitecting Legacy Machine Learning Systems
Amit Jain, Machine Learning Team Lead, TradeRev

  • Abstract: TradeRev uses regression models for predicting the auction price of cars. The early years of ML/development focused entirely on time to market which lead to a successful product but we ended up with a code base that had huge tech debt (spaghetti code, monolithic architecture, manually created infrastructure etc.). Increasing adoption rate of the product exposed the tech debt as scaling the product became a massive bottleneck.

    The speaker will discuss how they took the challenge of rearchitecting the entire ML product from both software engineering and data science perspectives.
    They will share how they accomplished many milestones as a result of this endeavour:

    - Improved model accuracy
    - Microservices architecture
    - Scalable ML solution
    - Continuous Integration & Automated Deployments
    - Dockerized software solution- 80% + code coverage
    - Regression/performance testing
    - Enhanced monitoring of evaluation metrics
    - Infrastructure as service

  • What you’ll learn: Attendees will learn tips, and techniques to embark on rearchitecting a legacy ML system from inception to production.

Advanced Technical Talk - A Flexible Framework for Entity Resolution, A Flexible Framework for Entity Resolution
Hoyoung Jang, Data Scientist, ThinkData Works, Cheng Lin, McGill University

Abstract: A critical component of data management and enrichment pipelines is connecting large datasets from various sources to form a holistic view; to make connections between entities across data sources. Oftentimes, these entities – such as individuals, organizations, or addresses – may not have a unique identifier that can be used as a key to detect duplicates or to merge datasets on. ThinkData has developed a scalable entity resolution engine to solve these problems. After experimenting with both deep learning and traditional NLP techniques, the team has found the best balance of accuracy and performance. Specifically, we have achieved near-parity in accuracy compared to Magellan (the leading entity resolution project in research), albeit with much better performance metrics and greater scalability. This talk will discuss the importance of entity resolution, our approach to solving real-world challenges, and the potential in using entity resolution and graph relationships in tandem.

Applied ML Case Study Talk - Deep Reinforcement Learning at Zynga, Overcoming the challenges of using RL in production
Patrick Halina Software Architect/ML Engineering Manager, Zynga

  • Abstract: Deep Reinforcement Learning has seen a lot of breakthroughs in the news, from game playing like Go, Atari and Dota to self-driving cars, but applying it to millions of people in production poses a lot of challenges. The promise of Reinforcement Learning is automated user experience optimization. As one of the world’s largest mobile video game companies, Zynga needs automation in order to personalize game experiences for our 70 million monthly active users. This talk discusses how RL can solve many business problems, the challenges of using RL in production and how Zynga’s ML Engineering team overcame those challenges with our Personalization Pipeline.

  • Infrastructure discussed: Spark, TensorFlow, TensorFlow-Agents

  • What you’ll learn: How Reinforcement Learning can be applied to many business problems, the challenges of dealing with RL in production over traditional supervised models, Zynga's solutions for those challenges

Advanced Research Talk - Explain Yourself! Leveraging Language Models for Commonsense Reasoning
Nazneen Rajani Research Scientist, Salesforce Research

  • Abstract: Deep learning models perform poorly on tasks that require commonsense reasoning, which often necessitates some form of world-knowledge or reasoning over information not immediately present in the input. We collect human explanations for commonsense reasoning in the form of natural language sequences and highlighted annotations in a new dataset called Common Sense Explanations (CoS-E). We use CoS-E to train language models to automatically generate explanations that can be used during training and inference in a novel Commonsense Auto-Generated Explanation (CAGE) framework. CAGE improves the state-of-the-art by 10% on the challenging CommonsenseQA task. We further study commonsense reasoning in DNNs using both human and auto-generated explanations including transfer to out-of-domain tasks. Empirical results indicate that we can effectively leverage language models for commonsense reasoning.
  • What you’ll learn:
    - Human explanations used only during training improves performance on downstream tasks
    - Explanations are a way to incorporate commonsense in neural networks
    - Language Models are powerful enough to generate meaningful commonsense explanations
    - Auto-generated explanation improves accuracy by 10% points on Commonsense Question Answering.

Poster Sessions:

Gavin Weiguang Ding, Senior Researcher, Borealis AI

  • On the Sensitivity of Adversarial Robustness to Input Data Distributions (ICLR 2019)

Harris Chan, Graduate Student, Vector & UofT

  • ACTRCE: Augmenting Experience via Teacher's Advice For Multi-Goal Reinforcement Learning

Neda Navidi, ML researcher, AI-r

  • Human/AI interaction loop

Chundi Liu, Data Scientist Intern, Layer 6 AI (NIPS 2019)

Hoora Fakhrmoosavy, Researcher, Ryerson University

  • Application of Brain emotional learning (BEL) model in prediction
  • Hybrid PSO–parallel brain emotional learning inspired model for generating artificial earthquake records

Angus Galloway, PhD Student in Machine Learning, University of Guelph

  • Batch Normalization is a Cause of Adversarial Vulnerability (ICML 2019)

Farukh Jabeen, Research Scientist, Computation, Science Research and Development

  • Cheminformatic Approach-In silico discoveries of novel drug lIke molecules, polymers and coatings

Jonathan Lorraine, Graduate Researcher, Vector & UofT

  • Optimizing Millions of Hyperparameters by Implicit Differentiation

Peter Starszyk, Data Scientist, PeakPower

Sicong Huang, Undergrad Research Student, UofT

  • TimbreTron: A WaveNet (CycleGAN (CQT (Audio))) Pipeline for Musical Timbre Transfer (ICLR 2019)

Dr. Joseph Geraci, CEO, NetraMark Cor,

  • A novel way to Understand Precise Subpopulations for Smaller Patient Populations but with Many Variables

Paul Vicol, Graduate Student, Vector & UofT

  • On the Invertibility of Invertible Neural Networks

What to expect at TMLS;

Business Leaders, including C-level executives and non-tech leaders, will explore immediate opportunities, and define clear next steps for building their business advantage around their data.

Practitioners will dissect technical approaches, case studies, tools, and techniques to explore challenges within Natural Language Processing, Neural Nets, Reinforcement Learning, Generative Adversarial Networks (GANs), Evolution Strategies, AutoML and more.

Researchers will have the opportunity to share with their peer's cutting-edge advancements in the field.

Machine learning, deep learning, and AI are some of the fastest growing and most exciting areas for knowledge workers - simultaneously, they are the key to untapped revenue sources and strategic insights for businesses. Firms are using AI to create unprecedented business advantages that are reshaping the global - but more specifically Canadian - economic landscape. Practitioners are leveraging and expanding their expertise to become high-impact global leaders.

Despite the vast opportunities that lie within our data, there are also explicit challenges to revealing their potential. Furthermore, transitioning to a career in practicing AL/ML, or managing ML and AI-driven businesses, are less than straightforward.

Why should I attend Toronto Machine Learning Society (TMLS) : 2019 Annual Conference & Expo

Developments in the field are happening fast - for practitioners, it's important to stay on top of the latest advances. Business leaders know that the implementation of new technology brings specific challenges.

The goal of TMLS is to empower practitioners and business leaders with direct contact to the people that matter most. For data practitioners, you'll hear how to cut through the noise and find innovative solutions to technical challenges. Business leaders will learn from the experience of those who have successfully implemented ML/AI and actively manage data teams.

Seminar series content will be practical, non-sponsored, and tailored to our local ecosystem. TMLS is not a sales pitch - It's a connection to a deep community that is committed to advancing ML/AI and to create and deliver value and exciting careers for Businesses and Individuals.

We're committed to helping you get the most out of the TMLS.

Joining together under one roof will be:

  • Machine Learning/deep learning PhDs and researchers

  • C-level business leaders

  • Industry experts

  • Enterprise innovation labs seeking to grow their teams

  • Community and university machine learning groups

TMLS tickets give access to the entire portfolio of:

  • Workshops

  • Real case studies

  • Keynote addresses

  • Poster Sessions and Vector Institute's Best Poster Award

  • Women in Data Science Evening event

  • Start-up Career Fair evening

  • High-level and granular discussions

  • Ample networking opportunities

  • Tailored post-event gatherings

  • Community follow-ups through a social channel.

Site: www.torontomachinelearning.com

Steering Committee & Team

Who Attends


Q: Are there ID or minimum age requirements to enter the event?

There is not. Everyone is welcome.

Q: Is food served?

Yes, light breakfast, coffee and lunch are served both days, catered by Oliver & Bonacini Restaurants

Q: Can I get a training certificate?

Yes, we can provide this upon request.

Q: Will tickets include access to the after-party?

Yes, attendees will have full access to both night's post-event networking socials.

Q: What are the transportation/parking options for getting to and from the event?

There are multiple parking options around College and Yonge, as well as the College Subway station and both the Yonge St Bus and College Streetcar.

Q: What can I bring into the event?

Just your ticket! You can also bring a CV if you're job seeking. Workshops will be the day before (19th) and we'll have separate instructions based on each workshop.

Q: How can I contact the organizer with any questions?

Please email info@torontomachinelearning.com

Q: What's the refund policy?

Tickets are refundable up to 30 days before the event.

Q: Why should I attend the TMLS?

Developments are happening fast - it's important to stay on top.

For businesses leaders, you will have direct contact with the people that matter most; consultants and experts, potential new hires, and potential new clients. For data practitioners, you'll have an opportunity to fast-track your learning process with access to relevant use-cases, and top quality speakers and instructors that you'll make lasting connections with while building your network.

The event is casual and tickets are priced to remove all barriers to entry. Space, however, is limited.

Q: Who will attend?

The event will have three tracks: One for Business, one for Advanced Practitioners/Researchers and one for applied use-cases (Focusing on various Industries). Business Executives, PhD researchers, Engineers and Practitioners ranging from Beginner to Advanced. See Attendee Demographics and a list of the Attendee Titles from our past event here.

Q: Will you focus on any industries in particular?

Yes, we will have talks that cover Finance, Healthcare, Retail, Transportation and other key industries where applied ML has made an impact.

Q: Can I speak at the event?

Yes you can submit an abstract here. Deadline to submit a talk is Sept 16th, however, we will continue to review submissions.

*Content is non-commercial and speaking spots cannot be purchased.

Q: Will you give out the attendee list?

No, we do our best to ensure attendees are not inundated with messages, We allow attendees to stay in contact through our slack channel and follow-up monthly socials.

Q: Can my company have a display?

Yes, there will be spaces for company displays. You can inquire at info@torontomachinelearning.com.

Share with friends

Date and Time


The Carlu

444 Yonge Street

Toronto, ON M5B 2H4


View Map

Refund Policy

Refunds up to 30 days before event

Save This Event

Event Saved