The top laptop here has served as my reliable backup since 2015 (got new in 2012) and will now once again step in as a replacement for yet another dead successor (bottom). This thing has filled in for 6 dead laptops of mine and others' over the years. Insane
Calling for applications for Carpentries Instructor Training Workshop organized by UofT. An immersive 2-day virtual workshop focusing on the pedagogy of teaching Coding & Data Science skills to novice learners. π§βπ«π§βπ»
Apply by tomorrow! See more in the link: https://t.co/G1zElOV3Rh
When you read a job description from a Big Tech company requiring production experience in deep learning and transformers...ππ
I guess they are hoping BERT's authors will apply π€·ββοΈπ€·ββοΈ
First big in-lab data collection day in over 2 years; I forgot how tiring but rewarding it is to gather your own data in a lab setting #research#AcademicTwitter
Reminder to my ML colleagues. @arxiv is not @github. Don't submit drafts to arXiv; you are wasting your colleagues' time. Submit when you have a final version ready for the research community to read and cite. Please enforce this rule on your students, too.
We are extremely excited to introduce our 2021β2022 UHNPA Leadership Committee! Welcome aboard!ππ
We will do our best to create a better community for our postdocs at UHN. Please stay tuned for our future events and opportunities!
#TeamUHNPA#postdoc#UHN
Frustrating that Ontario pays lip service to "being green", and then proposes projects like this that both encourage low-density transportation in the long term (of people and of goods), and will also produce substantial near-term carbon emissions #onpoli
https://t.co/wGHcSJwVTi
Day 1 of learning about coding #iOS apps using #SwiftLang was a success! Of course this is just a publicly available demo π but a start is a start! #congratslikeim5
Me thinking about a programming project in Python on Windows:
No problems, everything's fine here
Me thinking about a programming project in Swift on MacOS:
Explaining programming to a 6-year-old:
1. You have something you want to do.
2. You write code to do it.
3. The code doesn't work.
4. You fix the mistakes.
5. When the program works, you realize your idea was wrong.
6. You fix the idea.
7. Goto 2.
Itβs well known that neural networks model correlation and not causation.
Recently, Iβve found it helpful to think about NN blocks as literal correlations of correlations of correlations β¦
(incl. dense, norm, nonlin, conv, softmax, transformer, LMs, GANs, β¦)
π§΅ 1/20
Really interesting intersection of #DeepLearning and #Neuroscience. We still obviously have a very incomplete picture of the complexity of the squishy bits in our heads, but this is still a really neat thought-provoking line of work https://t.co/wZXYCYx22Q