Frank Xi

Devote my efforts to making progress in signal processing and machine learning

[Frank Xi @ChaosXi]

Spent an inspiring week at #ICASSP2026 in Barcelona. Great to reconnect with old friends and meet many brilliant researchers from around the world.
Also truly enjoyed the city — the architecture, food, and unique Mediterranean vibe made the whole experience even more memorable.

[Frank Xi @ChaosXi]

还记得你何时加入 X 吗？我知道！#我的X周年纪念日

[Greg Brockman @gdb]

GPT-5 Pro for catching subtle errors in academic work:

[Frank Xi @ChaosXi]

@irmohkle SORA 2

[Frank Xi @ChaosXi]

#ICASSP2024 opening ceremony@ICASSP2024

[IEEE ICASSP @ieeeICASSP]

A few highlights from the Welcome Reception!

[Julián Tachella @TachellaJulian]

📢📢 Release of DeepInverse library 📢📢

After months of intense work, we are releasing the first stable version of DeepInverse https://t.co/o7LpxJJjdz, a PyTorch library for solving inverse problems with deep learning.

with @HuraultSamuel, @MatthieuTerris and @ddongchen

A 🧵

[Frank Xi @ChaosXi]

Check out my photo in the IEEE Signal Processing Society's 75th Anniversary mosaic! #ICASSP2023 https://t.co/Dne6xBTbCA

[Frank Xi @ChaosXi]

Exciting to meet lots of friends #ICASSP2023

[Petros Boufounos @petrosb]

We are on!
#ICASSP2023 @ieeeICASSP

[Frank Xi @ChaosXi]

My paper was accepted to the 2023 IEEE International Conference on Acoustics, Speech & Signal Processing (@ieeeICASSP). Join me! https://t.co/BBswUhAq4p

[Aviation @xAviation]

F/A-18 Carrier landing - cockpit view ✈️

📸 Pundit https://t.co/eEC7DoLHK3

[Frank Xi @ChaosXi]

@wustlcig @ukmlv The arxiv link is not the right one.

[Computational Imaging Group (CIG) @uwcig]

Our last work "Deep Model-Based Architectures for Inverse Problems under Mismatched Priors" investigates the influence of mismatched CNN priors on image formation using deep model-based architectures.
✩ Project Page: https://t.co/xmwOn7ZGBc
✩ Paper: https://t.co/V6SM8NOoyw

[Kevin Patrick Murphy @sirbayes]

I am delighted to announce that the camera-ready version of my new book, "Machine Learning: Advanced Topics", is finally available online for free at https://t.co/XjR3nnLtOI (@mitpress will publish the hard copy in 2023.)

[elvis @omarsar0]

🎓 Introduction to Transformers

Excited to release my first lecture notes for the "Transformers United" course by Stanford. It includes a more beginner-friendly introduction to attention and transformers.

More coming soon!

https://t.co/pRPxw8iaHM

[Frank Xi @ChaosXi]

Batch normalization in 3 levels of understanding by Johann Huber in @TDataScience https://t.co/SzLGAJPV3J

[elvis @omarsar0]

Few-shot learning (FSL) methods like GPT-3 and Flamingo already show potential in handling diverse and new tasks.

This survey paper provides a nice summary of FSL, including applications & opportunities. This is an area of ML to watch closely.

https://t.co/DYFZHDvO8V

[Santiago @svpino]

A process to learn Machine Learning:

- Start with Python
- numpy, pandas, matplotlib
- Learn to use notebooks
- Find a (solved) problem
- Focus on the analysis, not the code
- Incorporate new algorithms as needed
- Look into the general ML process to approach problems
- Repeat

[Robert Lange @RobertTLange]

🎄I am a big fan of @ylecun‘s & @alfcnz‘s Deep Learning course. The attention to detail is incredible and one feels the love and passion, which goes into every single course week (my favorites: 7+8 on EBMs)🤗#feelthelearn

📜: https://t.co/1VgchQt2MI
📽️: https://t.co/LIZJP8NiMp