Reinforcement Learning

A good post by Alex Williams about Optimal Transport. alexhwilliams.info/itsneuronalblo…

Idk why, but as I remember, Python has package manager issues. When the packages are getting bigger, like vllm, the developers have no idea what should they consider in new versions deployment. We need to fix this!

I read some research papers - and then I'm like "This could've been a tweet" 😑

.Masoud presented the first beacon defense mechanism which can adapt against evolving membership inference attacks, using a reinforcement learning-based method. w/ Sobhan Shukueian, Kerem, Erman Ayday & Sinem Sav. doi.org/10.1101/2024.1…

#teambilkent RECOMB Conference Series

🚨 Exciting news! Our paper "Generated Data with Fake Privacy: Hidden Dangers of Fine-tuning Large Language Models on Generated Data" was accepted to USENIX Security '25 🎉🎉🎉 usenix.org/conference/use…

See our new paper with Sinem Sav and Kousar Saleem about a new vulnerability of the genomic data sharing beacon protocol in Bioinformatics👇

Such a great day to be an Iranian student!

Someone recommended me reading "think and grow rich". I'm on page 40, and still, it keeps saying that it is a very good book. So, what? I am reading you because I know you are good. I think it should be a trick or something. 🤔

I think I've just broken my therapist by explaining her how thinking works

High quality meme.

Over the weekend, I built a workflow using n8n.io. It goes over the internet and finds PhD positions in the fields similar to mine. It grabs everything with even a bit of similarity to my field. Normally, if I'd wanted to code something like this, it would've taken weeks!

margin: -5em

I think they started copying from xiaomi in retaliation...

Just updated my Mac. What the hell is this!??? Apple