Kaggle

Check out ‪@robmulla‬'s live data analysis of our 2022 annual user survey results!

2 years ago | [YT] | 26

Kaggle

Did you miss this episode of Chai Time ft. the Kaggle Staff? We're a little biased, but it's pretty good! Thanks to ‪@ChaiTimeDataScience‬

4 years ago | [YT] | 9

Kaggle

We love this livestream by Kaggle GM ‪@abhishekkrthakur‬

5 years ago | [YT] | 11

Kaggle

No streaming this Wednesday due to the holiday, but we'll be back on Friday at the usual time. Talk to you then and warm wishes for the holidays. :)

5 years ago | [YT] | 35

Kaggle

No live streaming this morning; Rachael is still sick. 😷😞

6 years ago | [YT] | 40

Kaggle

No stream today, Kagglers. Rachael is out sick. 😞

6 years ago | [YT] | 36

Kaggle

No livestreaming this week, Kagglers! Rachael is traveling all week.

We'll see you next Wednesday, Nov 20, at the usual time. 😊

6 years ago | [YT] | 52

Kaggle

Time to pick the next paper for the Kaggle reading group! Here are your choices:

> "Learning from Dialogue after Deployment: Feed Yourself, Chatbot!" (Hancock et al, 2019) was published at
ACL in 2019 & proposes a method for a dialog agent to continue to learn and update with user feedback. aclweb.org/anthology/P19-1358.pdf> "AutoML using Metadata Language Embeddings" (Drori et al, unpublished) proposes a method to augment existing AutoML systems by incorporating information from the text description of the dataset & description of possible algorithms to use. arxiv.org/pdf/1910.03698.pdf> "Machine Learning for Scent: Learning Generalizable Perceptual Representations of Small Molecules" (Sanchez-Lengeling et al, unpublished) outline an approach to predicting the scent of a molecule given it's structure using graph neural networks. arxiv.org/pdf/1910.10685.pdf

6 years ago | [YT] | 32

Kaggle

Alright folks, time to vote on the next paper for the Kaggle reading group! 📑🤓 Your options are:


* EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks (Tan & Le 2019) is from ICML 2019 and proposes a new method for scaling up CNN models & improving accuracy. proceedings.mlr.press/v97/tan19a/tan19a.pdf* ALBERT (anonymous, under review) is "a new pretraining method that establishes new state-of-the-art results on the GLUE, RACE, and SQuAD benchmarks while having fewer parameters compared to BERT-large." openreview.net/forum?id=H1eA7AEtvS* Deep Learning for Symbolic Mathematics (anonymous, under review) "train[s] a neural network to compute function integrals, and to solve complex differential equations". openreview.net/forum?id=S1eZYeHFDS

6 years ago | [YT] | 22

Kaggle

No live coding on Friday because Rachael is Out of Office. She'll be back next Wednesday for the Kaggle Reading Group, though; keep your eyes peeled for the poll for the next paper!

6 years ago (edited) | [YT] | 66