Chloe Robotics @ChloeRobotics

program a chatbot by having a conversation chloerobotics.com San Francisco, CA Joined November 2019

Tweets

253
Followers

145
Following

521
Likes

387

Carson 🧢 ⬆️ @Carsonlam

3 years ago

If you are scientifically minded and you feel uncomfortable bringing up the inconsistencies between science and religion with your religious friends, then you are allowing your friends to be a small part in the growing power of the religious lobby in our government

0 1 3 0 0

Carson 🧢 ⬆️ @Carsonlam

3 years ago

pubmed.ncbi.nlm.nih.gov/35704370/ When you train your neural network on many related tasks (#EHR time-series prediction tasks) simultaneously, it trains faster and improves the AUROC #medicalinformatics #multitasklearning #deeplearning #machinelearning #healthcare #datascience

0 5 7 0 0

Yann LeCun @ylecun

3 years ago

Worth repeating.

Brian Gitt @BrianGitt

3 years ago

Worth repeating.

2K 10K 38K 0 4K

Download Image

63 92 651 0 57

Carson 🧢 ⬆️ @Carsonlam

3 years ago

In this paper, the authors demonstrate positive transfer of skills across related medical time-series prediction tasks, with neutral to negative transfer to unrelated tasks while improving convergence time. #multitasklearning #NeuralNetworks

JMIR Publications @jmirpub

3 years ago

1 3 3 0 1

Download Image

0 1 1 0 0

JMIR Publications @jmirpub

3 years ago

New JMIR MedInform: Multitask Learning With Recurrent Neural Networks for Acute Respiratory Distress Syndrome Prediction Using Only #ehr Data: Model Development and Validation Study dlvr.it/SSDzYV

1 3 3 0 1

Download Image

/MachineLearning @slashML

4 years ago

I made the kind of tutorial I wish someone had made for me when I first started trying to connect the math in research papers with code examples I found online reddit.com/r/MachineLearn…

2 107 714 0 409

Chloe Robotics @ChloeRobotics

4 years ago

i made the kind of tutorial i wish someone had made for me when I first started out in machine learning github.com/clam004/intro_… #DeepLearning #lifelonglearning #ML

0 1 0 0 0

Carson 🧢 ⬆️ @Carsonlam

4 years ago

@Brian_FP_81 @CarGoHelmet @ezraklein @ATabarrok Everybody is self interested but some are more self interested than others, and when a society's average leader, in politics or business is a JD or MBA rather than a PhD or MPH, it tips the scale to the side of self interest via beauracracy rather than efficiency and innovation

0 1 2 0 0

Lex Fridman @lexfridman

4 years ago

Here's my conversation with Tim Urban (@waitbutwhy) author of the blog Wait But Why and one of the most fascinating, insightful, and fun humans I have the pleasure of knowing and learning from. His blog is a national treasure. youtube.com/watch?v=0Jd7fJ…

139 342 5K 0 357

Download Image

Carson 🧢 ⬆️ @Carsonlam

4 years ago

God: my people, imagine if the scientist that cured cancer had used that brain instead to find inefficiencies in the market so they could make rich people richer and party with models Secretary: God, pssst, eh hem, this IS one of those universes God: oh shit . . . eh, sorry folks

0 1 2 0 0

Chloe Robotics @ChloeRobotics

4 years ago

Leaning San Francisco skyscraper is tilting 3 inches per year as engineers rush to implement fix nbcnews.com/news/us-news/l… too bad they're all software engineers over there

0 0 0 0 0

Carson 🧢 ⬆️ @Carsonlam

4 years ago

Check out and #peerreview this interesting #JMIR #Preprint: Multi-Task Learning with Recurrent Neural Networks for ARDS Prediction using only EHR Data: Model Development and Validation Study preprints.jmir.org/preprint/36202 #peerreview # digitalhealth via @jmirpub

0 2 3 0 0

JMIR Publications @jmirpub

4 years ago

Peer-review me: Multi-Task Learning with Recurrent Neural Networks for ARDS Prediction using only EHR Data: Model Development and Validation #Study dlvr.it/SGkPPK (preprint)

1 5 3 0 1

Carson 🧢 ⬆️ @Carsonlam

4 years ago

If an engineer understands a concept 95% they will say the don't fully get it. If a politician understands something 15% they will say they are an advisory expert on the topic.

3 21 97 0 1

Elon Musk @elonmusk

4 years ago

Better resolution

2K 13K 86K 0 12K

Download Image

Andrej Karpathy @karpathy

4 years ago

Actually the ATP Synthase (and proton gradients) is by far the coolest molecular invention of life, followed by the Ribosome and then maaaaybe DNA, despite it being so iconic and getting so much press. Tell your friends.

55 70 983 0 63

Richard Socher @RichardSocher

4 years ago

It is kind of shocking how much current computing architectures (graphics processing units) define neural network architectures. GPUs made models like Transformers fast and hence popular, not neuroscience or theory. Our model space is restricted to what's fast on GPUs.

39 129 1K 0 84

Andrej Karpathy @karpathy

4 years ago

Super excellent and exciting read and direction! 🔥💯👏 Explicitly thinking of neural nets as code ("Software 2.0" as I referred to it in an earlier post), and adapting all of our extensive and existing Software 1.0 ecosystem to this new programming paradigm.