Alexander Quessy

Ohh exciting, would be keen to help out, anything in particular you’re thinking of implementing?

14.10.2025 21:27 — 👍 0    🔁 0    💬 0    📌 0

I’m surprised Gemini doesn’t push post-training super hard too given deepminds presumably very strong RL infrastructure.

13.10.2025 22:29 — 👍 0    🔁 0    💬 0    📌 0

Grad programs often don't teach folks how to write good reviewer reports. Here are some resources:

1. @brendannyhan.bsky.social review checklist: thepoliticalmethodologist.files.wordpress.com/2016/02/tpm_...

2. "How to Write an Effective Referee Report": aeaweb.org/articles?id=...

What else?

11.08.2025 23:59 — 👍 110    🔁 31    💬 6    📌 2

Robotic false positives are the best

Some VLA work I've been doing.
Instruction: "pick up the black bowl between the plate and the ramekin and place it on the plate success"
(\pi_{0} on Libero)

06.08.2025 15:27 — 👍 1    🔁 0    💬 0    📌 0

Have you seen any approaches to rlhf that include hierachical/curriculum learnin? It would be interesting to see the trade offs with this 🤔

17.07.2025 20:36 — 👍 0    🔁 0    💬 0    📌 0

*Please repost* @sjgreenwood.bsky.social and I just launched a new personalized feed (*please pin*) that we hope will become a "must use" for #academicsky. The feed shows posts about papers filtered by *your* follower network. It's become my default Bluesky experience bsky.app/profile/pape...

10.03.2025 18:14 — 👍 510    🔁 293    💬 24    📌 79

∇Q Exploring Machine Learning, Robotics, and AI through projects, experiments, and technical notes by Alexander Quessy.

Full blog at: aos55.github.io/deltaq/
#robotics #machinelearning #reinforcementlearning

19.02.2025 16:26 — 👍 0    🔁 0    💬 0    📌 0

SAC PickAndPlace robot picking up cube, https://github.com/AOS55/RLFoundations

I've also open-sourced some code - SAC and IL implementations for Fetch Pick-and-Place, with pre-trained models on HF. If you're interested in contributing, the repo is open for PRs!

github.com/AOS55/RLFoun...

19.02.2025 16:26 — 👍 0    🔁 0    💬 1    📌 0

Currently covering:

- Why "simple" tasks are often hard for robots
- Learning from demonstrations
- Reinforcement learning in robotics
- Common supervised learning problems in robotics

Coming soon:

- Sim2real transfer
- Modern approaches
- Real-world application

19.02.2025 16:26 — 👍 1    🔁 0    💬 1    📌 0

∇Q, aims to break down why robotic learning is different from standard ML. Sure, we can train vision models on millions of images, but why can't we do the same with robots?

19.02.2025 16:26 — 👍 0    🔁 0    💬 1    📌 0

∇Q Exploring Machine Learning, Robotics, and AI through projects, experiments, and technical notes by Alexander Quessy.

I've just launched ∇Q, a blog on robotic learning!

aos55.github.io/deltaq/

A friend recently said he wanted to get into robotic learning but didn't know where to start. The field can be pretty overwhelming - it pulls from ML, control theory, computer vision, and more.

19.02.2025 16:26 — 👍 0    🔁 0    💬 1    📌 0