''The Number of People Predicting the Death of Moore's Law
Doubles every 2 years''
― VP Microsoft Research
''The future is already here – it's just not evenly distributed''
― William Gibson
•Top 0.1% (2th out of 1,600+ teams) - https://tinyurl.com/ux6ztrw - Reboot: Box-Plots for Education
•Top 0.4% (7th out of 1,700+ teams) - https://tinyurl.com/s2f3yxt - Richter's Predictor: Modeling Earthquake Damage
•Top 1.3% (37th out of 2,700+ teams) - https://tinyurl.com/y4czhn24 - United Nations Millennium Development Goals
•Top 5% (46th out of 800+ teams) - https://tinyurl.com/y5al6ch9 - Flu Shot Learning: Predict H1N1 and Seasonal Flu Vaccines
https://www.kaggle.com/jeromeblanchet/competitions
My long-run performance at Kaggle is a Sigmoid function
•Top 01% - Digit Recognizer - Learn computer vision fundamentals with the famous MNIST data
•Top 01% - Housing Prices Competition for Kaggle Learn Users (InClass 19k+ teams)
•Top 01% - House Prices: Advanced Regression Techniques
•Top 01% - Titanic: Machine Learning from Disaster
•Top 02% - Предсказать день визита - Соревнования по курсу Машинного обучения Александра Дьяконова
•Top 02% - Real or Not? NLP with Disaster Tweets
•Top 02% - Categorical Feature Encoding Challenge II (20th out of 1,161 teams)
•Top 04% - Mlcourse.ai: Flight delays - Predict whether a flight will be delayed for more than 15 minutes
•Top 05% - Catch Me If You Can - Intruder Detection through Webpage Session Tracking
•Top 05% - Predict Future Sales
•Top 05% - Adult-PMR3508
•Top 11% - Zillow Prize: Zillow’s Home Value Prediction (Zestimate)
•Top 13% - M5 Forecasting - Accuracy - Estimate the unit sales of Walmart retail goods
•Top 17% - Flower Classification with TPUs
•Top 20% - M5 Forecasting - Uncertainty - Estimate the uncertainty distribution of Walmart unit sales
•Top 22% - Santa's Workshop Tour 2019
•Top 22% - Personalized Medicine: Redefining Cancer Treatment - Predict the effect of Genetic Variants
•Top 27% - Deep Fake Detection Challenge - Identify videos with facial or voice manipulations
Just starting exploring this new AI/ML competition platform https://www.aicrowd.com/participants/jerome_blanchet
•Kaggle: https://tinyurl.com/tfodmng
•Gmail: [email protected]
•AICrowd: https://tinyurl.com/ue3bodf
•LinkedIn: https://tinyurl.com/y99wzc5r
•DrivenData: https://tinyurl.com/ux6ztrw
•Twitter: https://twitter.com/JrmeEBlancht1
•InfoNex: https://infonex.com/1382/
•LifesonAI (Oganization): https://github.com/LifesonAI
•GitHub (personal account): https://github.com/JeromeBlanchet
•Quora: https://www.quora.com/profile/J%C3%A9r%C3%B4me-E-Blanchet
•Stack Overflow (I joined recently): https://stackoverflow.com/users/13966273/j%c3%a9r%c3%b4me-blanchet?tab=profile
•Facebook AI & Deep Learning Memes Moderator: https://www.facebook.com/groups/1638417209555402/members/admins
https://www.kaggle.com/jeromeblanchet/discussion
Net Votes: 75
Nb Votes/Nb Post: 6.25
•Dimension Reduction for Predictive Modelling and Clustering (41 x up-voted)
•LSTM Neural Network & Dropout Regularization Strategy (9 x up-voted)
•Concatenating PCA & Correspondence Analysis Factors in your modelling (6 x up-voted)
•Building Innovative Predictor for Financial Crisis (5 x up-voted)
•Montreal, Theano & Yoshua Bengio (4 x up-voted)
•Dendrogram and Hierarchical Clustering Logic (3 x up-voted)
•How to detect the Noisy Component of your PCA Factors? (3 x up-voted)
•Approaches for Handling Missing Data (2 x up-voted)
•Self-organizing Map (SOM) (1 x up-voted)
•The FICO Competition Challenge is about Interpreting Black Box Algorithms
https://www.kaggle.com/jeromeblanchet/datasets?sort=votes
Mainly in the field of Natural Language Processing & Question Answer Learning.
Most of these datasets have their own BERT Leaderboard, & represent a technical variation of the SQuAD Dataset.
NLP Datasets:
•CommonsenseQA
•SciQ
•ReCoRD
•Spider 1.0
•SParC 1.0
•CoSQL 1.0
•DuoRC
•MultiRC
•ShARC
•SWAG
•ARC
•HotPotQA
•RecipeQA
•QuAC
•AQUA-RAT
•DROP
•QuaRTz
•CoQA
Other Datasets:
•Fannie Mae & Freddie Mac Database 2008-2018
•Johns Hopkins University’s Department of Computer Science Multi-Domain Sentiment Data
•Federal Home Loan Level Bank System 2009-2018
•Canada Civil Aircraft Fatalities & Injuries
Honored to be Speaker at the Artificial Intelligence for the Public Sector Conference 2021
Attendance Fee: $1,800-$2,400/person (free for speakers)
Main Page: https://infonex.com/1382/
Agenda: https://infonex.com/1382/agenda/
Biographies: https://infonex.com/1382/speakers/
PDF: https://infonex.com/1382/agenda-pdf/?pdf=Brochure
10 speaker, which include,
------Carter Cousineau (Ph.D), Managing Director, Centre for Ethical Artificial Intelligence, The University of Guelph
------Brian Drake, Director of Artificial Intelligence, Defense Intelligence Agency (DIA)
------Yvan Gauthier, Senior Defence Scientist, Department of National Defence
------Sevgui Erman (Ph.D), Director of Data Science, Statistics Canada
------Maryam Haghighi, Director of Data Science, Bank of Canada
My Table of Content for Canadian AI Conference 2021:
Recent Research & Development in Time Series Forecasting Models
------Uber’s Hybrid Method of Exponential Smoothing & Recurrent Neural Network (The ES-RNN Model)
------Element AI’s Neural Basis Expansion Analysis for Interpretable Time Series Forecasting (The N-BEATS Model)
------Facebook's Neural Network for Time-Series (The NeuralProphet Model)
Recent Research & Development in Natural Language Processing Models
------Google Brain’s Attention is all you Need Paper (The Transformer Model)
------Google AI’s Pre-training of Deep Bidirectional Transformers for Language Understanding (The BERT Model)
------OpenAI’s Generative Pre-trained Transformer (The GPT-1-2-3 Models)
Selected Bibliography for My Canadian AI Conference content:
•Fast ES-RNN: A GPU Implementation of the ES-RNN Algorithm https://arxiv.org/pdf/1907.03329.pdf
•N-BEATS: Neural Basis Expansion Analysis for Interpretable Time Series Forecasting https://arxiv.org/pdf/1905.10437.pdf
•AR-Net: A Simple Auto-Regressive Neural Network for Time-Series https://arxiv.org/pdf/1911.12436.pdf
•Attention Is All You Need https://arxiv.org/pdf/1706.03762.pdf
•BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding https://arxiv.org/pdf/1810.04805.pdf
•Language Models are Few-Shot Learners https://arxiv.org/pdf/2005.14165.pdf
•(2022-) Unit Head - Senior Data Science Engineer - Center for Special Business Projects - DEIL - at Statistics Canada
•(2021) Senior Data Scientist & Senior Economic Advisor - Data Science Division - at Transport Canada
•(2021) Speaker at Artifial Intelligence Conference https://infonex.com/1382/ Transformers, Deep Learning, Self-Attention, LSTM & NLP
•(2020) Senior Analyst Lead - Model Development (NLP), at Transport Canada, Ottawa, Ontario, Canada
•(2015-2019) Senior Specialist - Model Development & Housing Finance Stability Researcher, at CMHC, Ottawa, Ontario, Canada
•(2013-2014) Analyst - Model Development, at DDM Group, Québec City
•(2012) Academic Research Assistant in Econometric, Model Development, at Sherbrooke University
•(-2011) Bachelor's in Mathematics from Montreal University & Master's in Econometric from Sherbrooke University
•Neuraxio's Neuraxle Framework for Clean Deep Learning Pipeline: https://github.com/Neuraxio/Neuraxle/blob/master/README.rst
Please find below a couple of my public repository. I am currently working on several private project.
Please contact me for details about these private repositories.
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent