כמה תובנות מסדנאות לחינוך פיננסי בבתי ספר: 1. התלמידים עפים על זה! ברוב השיעורים שלי בחרו לדלג על ההפסקה. רוצים להבין עוד. איך עובד בנק, מה זה הבורסה ועל החשיבות של חסכון והשקעה 2. לדעתי, הדבר שהכי חשוב ללמד אותם זה איך עובדת ריבית ובפרט את הכח העצום שיש לריבית ד'ריבית לאורך זמן

Nothing binds people together like a common enemy

Woop Live at Salar de Uyuni for Cercle now on YT youtu.be/sCNlt5nvSI8 ⠀ half planed - half improvised

Learning embedding space for location?! Turns out the answer is... yes! 🙂 Two absolutely fascinating reads with a lot of good information on approaching DL projects in general. The first one is a blog post by Sentiance sentiance.com/2018/05/03/ven….

Blogpost on our new theory for word2vec-like representation learning methods for images, text, etc. Explains why representation do well on previously unseen classification tasks offconvex.org/2019/03/19/CUR… Relevant to meta learning, transfer learning? Paper arxiv.org/abs/1902.09229

neat dl for image recognition result: input representation does matter! (convert the RGB into diff color spaces, feed each to a diff network, combine with a linear layer, and viola: few params, strong results. ensemble of manually crafted representations) arxiv.org/abs/1902.00267

How many companies that use Google's ML engine have read the ToS? "Customer will not ... use these Services to create, train, or improve ... a similar or competing product or service ... such as cars, TVs, appliances, or speakers without Google's prior written permission"

Pro Project Jupyter tip I literally just learned today: If you need to undo *inside* a cell, hit CTRL + Z If you need to undo *outside* a cell (like for **un-deleting a cell you just deleted**) hit ESC + Z Retweet to save a live 🙏 (I've deleted so. many. cells. 😩)

This degree of user customization and lack of safeguards on display here is exactly what Axon's AI ethics board said should not be sold to police: x.com/VICE/status/11…

I'm working on a book on machine learning interviews so I've been spending the last few months talking to companies about their hiring process for ML roles. This thread is a summary of what I've learned. It will be updated as the book progresses. (1/n)

What's Hidden in a Randomly Weighted Neural Network? “Hidden in a randomly weighted Wide ResNet-50 we show that there is a subnetwork (with random weights) that is smaller than, but matches the performance of a ResNet-34 trained on ImageNet.” 😮 arxiv.org/abs/1911.13299

(5/5) 5. מכון ויצמן למדע קורא למשרד הבריאות להמשיך ולפעול יחד, בשותפות אמת ובהגינות, במאבק הלאומי בקורונה, לטובתם ולרווחתם של כל תושבי ישראל. התחמקות מאחריות ו"חיפוש אשמים", מקומם לא יכירם בתחומי המדע והרפואה.

This is deeply wrong Predicting students future grades from school history + past grades *and* making automatic decisions based on them is the exact example of a system that should *not* be deployed And the fact that it already determined the fate of 170k students is horrifying

Let's do a giveaway 🎁 !!! 1⃣ A full year subscription to Real Python ($199) 2⃣ The Good Parts of AWS (taught by Daniel Vassallo) ($35) 3⃣ A full year subscription to Pluralsight ($199) That's $435! To participate: ▫️Follow me ▫️Like this tweet ▫️Retweet it 🧵👇

This is only the tippity tip of the iceberg. There are literally dozens of examples like this. One reporter got access to our mental health channel on Slack and started reaching out to the students who had talked about suicide, asking for comment. Imagine how that must feel.

1. שאול, בעל אטליז בדרום, היה תחת פיקוח וטרינרי של הרשות המקומית שלו. החל מהשנה הוא ישלם על אותו שירות בדיוק 13,000 שקלים בחודש. פיקוח זה טוב אבל למה צריך לשלם עשרות אלפי שקלים בחודש? כי המדינה החליטה להקים עוד תאגיד, שיוסיף עוד בירוקרטיה ועוד עלויות עבור אותו פיקוח בדיוק

⚡New Handle: @LightningAI ⚡ fast.ai After diving in...wtf? Where's the credit? It's obviously ripped from fastai. I mean it's not just the library, they even use the same marketing lines. As a heavy pytorch lightning user, I don't think I can use libraries made by that group anymore in good conscience :(

The founder of Grid.ai () has been sharing some falsehoods about fastai as he promotes the pytorch lighting library. I want to address these & to share some of our fast ai history. 1/

Personal Thoughts: (not for everyone, feel free to skip) For the longest time, I struggled with my identity. A Palestinian kid born inside Israel. Like…wtf. Many of my friends refuse to this day to say the word “Israel” and call themselves “Palestinian” only. But since I