• Machine Intelligence Toronto

    MaRS Centre

    PAPER REVIEW: Estimation and Inference of Heterogeneous Treatment Effects using Random Forests (Wager and Athey, July[masked]) https://arxiv.org/pdf/1510.04342.pdf Format 18h30 - Welcome, Thank You, Thank you Georgian Partners, Autodesk 18h32 - Who is hiring? 18h35 - Let's discuss Wager and Athey 2017 because it is so hot right now 19h30 - Networking / Talk to the people who are hiring 19h55 - You don't have to go home but you can't stay here Note the new location. We're upstairs now, at the Autodesk space next to the starbucks. Thank you to our friends at Georgian Partners for sponsoring the space with Autodesk!

    7
  • Machine Intelligence Toronto

    MaRS Centre

    Public Media, Social Cohesion, and the Filter Bubble Christopher Berry, Director Product Intelligence, CBC Abstract: Public media produces a public good in the form of social cohesion. Filter bubbles erode social cohesion. Recently, filter bubbles have grown stronger, more profitable, and more hackable. Filter bubbles, in part generated by deep learning algorithms, can be addressed with deep learning algorithms, and there are novel managerial implications for public media. Bio: Christopher is a data scientist. He turns data into product. He does it at the Canadian Broadcasting Corporation where he leads the product intelligence team. Previously, he founded authintic (sold to 500px in 2014), and led teams at Syncapse and Critical Mass. 30 minutes are set aside for chatting Note the new location. We're upstairs now, at the Autodesk space next to the starbucks. Thank you to our friends at Georgian Partners for sponsoring the space with Autodesk!

    5
  • Machine Intelligence Toronto

    MaRS Centre

    Epsilon: differentially private machine learning Mads Mihailescu, CTO, Georgian Partners @madumix https://georgianpartners.com/introducing-epsilon-v1-0/ Bio Madalin Mihailescu is Chief Technology Officer and a Partner, and is responsible for leading the Georgian Impact team. In addition, he provides technology architecture support to our portfolio companies. Madalin is also responsible for analyzing technology trends, engaging in the technology due diligence process and developing internal software to support our investment process. [Note the new location. We're upstairs now, at the Autodesk space next to the starbucks.]

    5
  • Machine Intelligence Toronto

    MaRS Centre

    What are Capsule Networks? Pavan Mirla - What is Capsule Network able to learn? - What are Capsule Functions? - What is Capsule Networks composed of? - What information does orientation of the output vectors encode? - What is role convolutional layers in Capsule Net ? - What is application of a Squashing function? - What do first/primary layer capsules trying to predict? - What is routing by agreement? - How to find cluster of agreement? - What is benefit of using margin loss instead of Cross Entropy? - What is safe approximation. How to address situation of zero vector? - What are digit capsules? - What is purpose of using Decoder? - What is reconstruction loss? BIO: Pavan Mirla's professional background includes over 8 years as Quant analyst/Data Scientist at Canada's Institutional Investors - Manulife Asset Management and CPP Investment Board. He held Sr. Data scientist position at Manulife/John Hancock Innovation lab.His passion is to learn and teach scientific topics - that includes Machine Learning and Quantitative Analysis. His technical interests are Optimization algorithms, Natural Language Processing, Knowledge Graphs/ Semantic Web and Recommender Engines. His ambition is to empower small/mid sized companies with AI capabilities and equip them with resources to compete with bigger firms. About #miToronto Both audiences, those who are interested in machine intelligence / data science, and those who are practitioners of machine intelligence / data science are invited, and can both expect to learn something new. Attendees can expect to learn what machine intelligence is, its applications, and what's going on in Toronto's data science community. Significant getting to know you time, and Q&A time is deliberately set aside. MARS AUDITORIUM, DOWNSTAIRS, near the Tim Horton's, use the glass doors (the unmarked wood ones are all locked). • What to bring • Important to know

    29
  • Machine Intelligence Toronto

    MaRS Centre

    Machine learning with fully anonymized data Adam Drake (@aadrake (https://twitter.com/aadrake)) Abstract Sharing data for any purposes, including for machine learning, is fraught with problems related to ethics, organizational policy and dynamics, and regulatory restrictions. In this talk I will discuss the topic of feature hashing and how, with a slight modification from the standard approach, we can satisfy many ethical and other requirements by training our models without any knowledge of the underlying data. In addition to presenting the ethical and regulatory benefits of building machine learning models using completely anonymized data, if time permits we may explore and discuss the performance characteristics and benefits of such an approach. Bio (also available at https://aadrake.com/press.html ): Adam Drake’s professional background includes a wide range of technical professional and management roles, including: leading technical business transformations in global and multi-cultural environments, performing in-depth technical due-diligence and funding analysis for investors, and mentoring new technical and operational executives. His passion is to help companies become more productive by improving internal leadership capabilities, and accelerating product development through technology and data architecture guidance. His technical interests include online learning systems, high-frequency/low-latency data processing systems, recommender systems, distributed systems, and functional programming. Adam has a background in Applied Mathematics and has worked in technology roles since the 90s. His career spans a variety of industries, including e-commerce, online travel, online marketing, financial services, healthcare, and oil and gas. LinkedIn: https://linkedin.com/in/aadrake (https://clicktime.symantec.com/a/1/D_A2ksaRfcJBO3yvLh1H5gK8QrSyZQjmnv35kCda4-Q=?d=y1bkEp-RU1DVgixrYxzART_TIu_3LFaJHY-gWwkFlvNVdWhIxuadDxvdhujk_rQ_PL8W2urGSAEfyHkFf8Oxndfw4hwOT6knLBRZPt_eps9GRufow5U3ajpw4dC09qJmXWTHFN0-formLSQupPkmyYe_sam-109vyc8wiw3PAvRRuX_S3hKz-FXLyrgLYSmZHlEKLLQyW29ulkN8oGsS-ynbFP6jgvxOFn8e8y5hvqzIlaooNpoxShVmU_tt17OdCmQwXP6y82rNo4kqyaKDjaSZcjs2QwI7Ei3PJqHWGq9xP0WQFlRj-XLi5t9eskgnEEx_UOttC_g6DtZ3cakI1OVZAc6QCCnd_bUQvdo6TsuBDMn6iG8F8aAsKszhuf8lWcCL5-ZG-vLuV9nRxjY%3D&u=https%3A%2F%2Flinkedin.com%2Fin%2Faadrake) Website: https://aadrake.com (https://clicktime.symantec.com/a/1/fsavZQkSgGmopFSZuXdqgfhOrDYwkuJ7GjvrVoIxWHA=?d=y1bkEp-RU1DVgixrYxzART_TIu_3LFaJHY-gWwkFlvNVdWhIxuadDxvdhujk_rQ_PL8W2urGSAEfyHkFf8Oxndfw4hwOT6knLBRZPt_eps9GRufow5U3ajpw4dC09qJmXWTHFN0-formLSQupPkmyYe_sam-109vyc8wiw3PAvRRuX_S3hKz-FXLyrgLYSmZHlEKLLQyW29ulkN8oGsS-ynbFP6jgvxOFn8e8y5hvqzIlaooNpoxShVmU_tt17OdCmQwXP6y82rNo4kqyaKDjaSZcjs2QwI7Ei3PJqHWGq9xP0WQFlRj-XLi5t9eskgnEEx_UOttC_g6DtZ3cakI1OVZAc6QCCnd_bUQvdo6TsuBDMn6iG8F8aAsKszhuf8lWcCL5-ZG-vLuV9nRxjY%3D&u=https%3A%2F%2Faadrake.com) Twitter: https://twitter.com/aadrake (https://clicktime.symantec.com/a/1/NXtno5vfoPw5MpqKpOreb8yIMsDCM9-9xd0EGuwTAgg=?d=y1bkEp-RU1DVgixrYxzART_TIu_3LFaJHY-gWwkFlvNVdWhIxuadDxvdhujk_rQ_PL8W2urGSAEfyHkFf8Oxndfw4hwOT6knLBRZPt_eps9GRufow5U3ajpw4dC09qJmXWTHFN0-formLSQupPkmyYe_sam-109vyc8wiw3PAvRRuX_S3hKz-FXLyrgLYSmZHlEKLLQyW29ulkN8oGsS-ynbFP6jgvxOFn8e8y5hvqzIlaooNpoxShVmU_tt17OdCmQwXP6y82rNo4kqyaKDjaSZcjs2QwI7Ei3PJqHWGq9xP0WQFlRj-XLi5t9eskgnEEx_UOttC_g6DtZ3cakI1OVZAc6QCCnd_bUQvdo6TsuBDMn6iG8F8aAsKszhuf8lWcCL5-ZG-vLuV9nRxjY%3D&u=https%3A%2F%2Ftwitter.com%2Faadrake) About #miToronto Both audiences, those who are interested in machine intelligence / data science, and those who are practitioners of machine intelligence / data science are invited, and can both expect to learn something new. Attendees can expect to learn what machine intelligence is, its applications, and what's going on in Toronto's data science community. Significant getting to know you time, and Q&A time is deliberately set aside. MARS AUDITORIUM, DOWNSTAIRS, near the Tim Horton's, use the glass doors (the unmarked wood ones are all locked).

    24
  • Machine Intelligence Toronto

    MaRS Centre

    How to implement AI in your software solution Parinaz Sobhani, Director, Machine Intelligence, Georgian Partners Twitter: @PariAIML (https://twitter.com/PariAIML) Website: https://georgianpartners.com/team/parinaz-sobhani/ Abstract Artificial intelligence (AI) is rapidly moving out of the laboratory and into business and consumer applications. As a result, we’re seeing a fundamental shift in how software is built and what it's capable of doing. In this talk, I will introduce the principles of applied artificial intelligence, along with a framework and maturity model for applying AI to organizations. These principles bring structure to an important, and at times highly complex topic, and can be a useful reference point when it comes time to develop and execute an AI strategy. Finally, I will talk about ethical problems, most importantly, bias and fairness in AI systems and interpretability of machine learning models and how to address these issues. Bio Parinaz Sobhani is the Director of Machine Learning on Georgian Impact and is responsible for engaging with portfolio companies to accelerate their growth using state-of-the-art artificial intelligence and machine learning techniques. Parinaz holds a Ph.D. from the University of Ottawa with a research focus on solving opinion mining problems using natural language processing and deep neural networks techniques. She has more than 10 years of experience in developing and designing new models and algorithms for various artificial intelligence tasks. Prior to joining Georgian Partners, Parinaz worked at Microsoft Research where she developed end to end neural machine translation models. Previous to this, she worked for the National Research Council of Canada, where she designed and developed deep neural network models for natural language understanding and sentiment analysis. About #miToronto Both audiences, those who are interested in machine intelligence / data science, and those who are practitioners of machine intelligence / data science are invited, and can both expect to learn something new. Attendees can expect to learn what machine intelligence is, its applications, and what's going on in Toronto's data science community. Significant getting to know you time, and Q&A time is deliberately set aside. MARS AUDITORIUM, DOWNSTAIRS, near the sad food court, use the glass doors (the unmarked wood ones are all locked).

    11
  • Machine Intelligence Toronto

    MaRS Centre

    The Joy of Designing Deep Neural Networks Bradley Arsenault, founder and CEO of Electric Brain (https://electricbrain.io/) Moderated by Jay Choi @torontosj (https://twitter.com/torontosj) Abstract Bradley Arsenault, founder and CEO of Electric Brain, will discuss how he discovered his passion for artificial intelligence and deep-learning. He will discuss the tremendous flexibility of deep neural networks and why they are more intuitively understand as graphs rather then equations. Finally, he will discuss some practical difficulties in translating research papers into practice. Bio Bradley Arsenault is a long time entrepreneur and computer programmer. He first started to get involved in artificial intelligence at his last company, Sensibill Inc, where he designed and developed an AI system for understanding and structuring images of receipts. Afterwards he moved onto Electric Brain, where he develops custom AI technologies for clients using deep-learning technology. Both audiences, those who are interested in machine intelligence / data science, and those who are practitioners of machine intelligence / data science are invited, and can both expect to learn something new. Attendees can expect to learn what machine intelligence is, its applications, and what's going on in Toronto's data science community. Significant getting to know you time, and Q&A time is deliberately set aside. MARS AUDITORIUM, DOWNSTAIRS, near the sad food court, use the glass doors (the unmarked wood ones are all locked).

    21
  • Machine Intelligence Toronto

    MaRS Centre

    The Top 10 Core Competencies of a Machine Intelligence Data Scientist Have your say: https://goo.gl/forms/XMYCgOTA6Ydvi6SH2 With summary and commentary by Christopher Berry and Jay Choi Abstract On November 8, 2016, the Machine Intelligence Community outlined the top 10 core competencies of machine intelligence data scientist. The objective of this survey is validate, moderate, and iterate upon that list. This survey should take around 20 minutes to complete. Obvious Personally Identifiable Information will be stripped prior to distribution. The dataset may be distributed publicly and it may be retained forever. Results will be presented and discussed at the September 14 #miToronto meetup. If you are new to Machine Intelligence, take the survey because you'd like to learn and have input. If you are familiar with Machine Intelligence, take the survey because there are gaps to be closed and to make the model better. Then come to #miToronto to hear what others said and discuss. Christopher and Jay will hold out until the day of before summarizing the results. There may be a WE'LL DO IT LIVE component. Have your say: https://goo.gl/forms/XMYCgOTA6Ydvi6SH2 Thank you to Georgian Partners (http://georgianpartners.com/) for the extra space! Both audiences, those who are interested in machine intelligence / data science, and those who are practitioners of machine intelligence / data science are invited, and can both expect to learn something new. Attendees can expect to learn what machine intelligence is, its applications, and what's going on in Toronto's data science community. Significant getting to know you time, and Q&A time is deliberately set aside. MARS AUDITORIUM, DOWNSTAIRS, near the sad food court, use the glass doors (the unmarked wood ones are all locked).

    11
  • Machine Intelligence Toronto

    MaRS Centre

    Title Differential privacy: Not just for academics and big tech Yevgeniy Vahlis Abstract Differential privacy is making headlines thanks to the pioneering work of companies like Apple and Google, and it is now being used by companies of all sizes to provide data privacy guarantees. It is no secret that machine learning models can memorize (overfit) training data and that through carefully crafted adversarial inputs machine learning models can be subverted by an attacker. Combine these facts with a model that aggregates data from a multitude of customers and you have an AI-driven disaster waiting to happen. In this talk we will cover a defensive measure called “differential privacy” that is a potential solution to such threats. In this talk Yevgeniy will explain the core concepts of differential privacy and share a behind the scenes look at how three leading SaaS companies are successfully implementing differential privacy in their products. About Yevgeniy Vahlis Yevgeniy Vahlis is the Director of Applied Research at Georgian Partners, a Toronto-based growth equity firm that invests in SaaS-based business software companies. Prior to joining Georgian Partners, he was a Senior Engineer at Amazon, where worked on deep learning applied to demand forecasting. Previous to that, Yevgeniy worked at Nymi, a biometrics security startup, and at AT&T’s Security Research Center. He holds a Ph.D. in cryptography from the University of Toronto. Thank you to Georgian Partners (http://georgianpartners.com/) for the extra space! Both audiences, those who are interested in machine intelligence / data science, and those who are practitioners of machine intelligence / data science are invited, and can both expect to learn something new. Attendees can expect to learn what machine intelligence is, its applications, and what's going on in Toronto's data science community. Significant getting to know you time, and Q&A time is deliberately set aside. MARS AUDITORIUM, DOWNSTAIRS, near the sad food court, use the glass doors (the unmarked wood ones are all locked).

    14
  • Machine Intelligence Toronto

    MaRS Centre

    Title Machine Intelligence Ethics Panel: Dwija Patel, Bart Gajderowicz, Matthew Kantor, Myles Harrison Abstract Why this topic? Because anxiety. The panel is tackling policy, employment, politics, cancer, self-driving cars, trolleys, Rand, reasoning around narrow machine intelligence. Lots of interaction time planned. Bio Dwija is a Senior Android developer at CBC and an initiator for women in tech at her organization. She is mainly involved in spreading awareness about technology in youth, she plays the role of a mentor and a coach for the next generation of technocrats. She has graduated in Information Technology. She has experience working in different environments, from start-up to banking to media and enjoys diverse work culture. Matt Kantor is VP engineering at MileOut, a product incubation lab launching from the founders of Normative. Matt has a love of technology, learning and design. He has been working in the industry for over 20 years. His career has been spent building teams that produce dependable, high-performance software He’s designed and built critical infrastructure for both Fortune 500 and startups. Matt can easily bridge the gap between line-level code and the 50 thousand foot architectural view. Redundant, scalable, and easy to use systems what makes Matt happy. Bart is a fourth-year PhD candidate in the Mechanical and Industrial Engineering Dept. (http://t.umblr.com/redirect?z=http%3A%2F%2Fwww.mie.utoronto.ca&t=ZjJiMGE1MDRkMWFiMTE3YzNiOWRlMzZhZjEwODNiMjQ2ZmY4MGM0NyxUbU4zdWpESg%3D%3D&p=&m=0)at the University of Toronto, under the supervision of Dr. Mark S. Fox (http://www.eil.utoronto.ca/members/msf/) and Dr. Michael Grüninger (http://stl.mie.utoronto.ca/gruninger.html). He is a member of the Centre for Social Service Engineering (http://csse.utoronto.ca/) working on the Social Service Simulator (http://csse.utoronto.ca/social-services-simulator) project. He holds a MSc and BSc degrees in Computer Science (http://cs.ryerson.ca/) from Ryerson University where he co-founded the Ubiquitous and Pervasive Computing Laboratory. Bart's current work applies artificial intelligence technique to understanding human behaviour, with a focus on evaluating social service policies and applications in "intelligence augmentation" and "human-computer collaboration". Prior to joining the CSSE, Bart spent 12 years working as a Software Engineer. Web: http://bartg.org Myles Harrison is an analytics consultant at Accenture. Prior to his current role, he has previously been employed in consulting at PwC and SapientNitro, as well as a data scientist in programmatic advertising. Myles has spoken on machine learning and data visualization at the Toronto Data Science Group, Toronto's Big Data Week and for the Toronto R User Group, and blogs on these topics as well as the analytics of everyday life at http://www.everydayanalytics.ca . Thank you to Georgian Partners (http://georgianpartners.com/) for the extra space! Both audiences, those who are interested in machine intelligence / data science, and those who are practitioners of machine intelligence / data science are invited, and can both expect to learn something new. Attendees can expect to learn what machine intelligence is, its applications, and what's going on in Toronto's data science community. Significant getting to know you time, and Q&A time is deliberately set aside.

    11