Minho Heo

@MinhoHeo

PhD student at KAIST | Robot Learning, Robotics

minoring.github.io Joined September 2022

168 Following 220 Followers 97 Posts

Latest posts 5

Minho Heo @MinhoHeo Sep 26, 2025

Come see what robot learning can do for surgical automation!
We’re excited to host the first Workshop on Automating Robotic Surgery with an amazing lineup of speakers.

🗓️ Sept. 27 09:30AM - 12:30PM
📍 Floor 3F, Room E7
🌐 …ng-robotic-surgery-workshop.github.io

#CoRL2025 #CoRL @corl_conf

Minho Heo @MinhoHeo Oct 4, 2024

Humans build a reusable skill library (e.g., grabbing) and adapt it to perform new tasks with specific variations (e.g., grabbing a cup, grabbing a dish). Surprisingly, we can equip robots with this same ability! Check out our paper, EXTRACT, where we delve into this topic!

Jesse Zhang ✈️ CoRL 2026 Verified account @Jesse_Y_Zhang Oct 3, 2024

How can robots efficiently learn **new tasks/in new settings**?

Introducing EXTRACT: a reinforcement learning (RL) framework that extracts a discrete + continuously parameterized skill library from offline data for efficient RL on new tasks!

Accepted to CoRL 2024: 🧵👇

View quoted post on X

Minho Heo @MinhoHeo Sep 5, 2024

Fine-tuning RL for long-horizon, complex tasks with sparse rewards is extremely challenging, but this work achieves remarkable results! It's exciting to see how this pushes the boundaries in furniture assembly tasks 😀

Allen Ren Verified account @allenzren Sep 4, 2024

👇Introducing DPPO, Diffusion Policy Policy Optimization

DPPO optimizes pre-trained Diffusion Policy using policy gradient from RL, showing 𝘀𝘂𝗿𝗽𝗿𝗶𝘀𝗶𝗻𝗴 𝗶𝗺𝗽𝗿𝗼𝘃𝗲𝗺𝗲𝗻𝘁𝘀 over a variety of baselines across benchmarks and sim2real transfer

diffusion-ppo.github.io

View quoted post on X

Minho Heo @MinhoHeo May 18, 2024

Great job on handling complex manipulation tasks with sim2real + minimal human help! Really loved the visuals, especially how the robot did the lightbulb assembly!

Yunfan Jiang Verified account @YunfanJiang May 17, 2024

Does your sim2real robot falter at critical moments 🤯? Want to help but unsure how, all you can do is reward tuning in sim 😮‍💨?

Introduce 𝐓𝐑𝐀𝐍𝐒𝐈𝐂 for manipulation sim2real. Robots learned in sim can accomplish complex tasks in real, such as furniture assembly.
🤿🧵

View quoted post on X

Minho Heo @MinhoHeo Jul 5, 2023

Looking for a challenging manipulation benchmark?

Introducing FurnitureBench 🪑🛠️, a reproducible real-world furniture assembly benchmark (#RSS2023 @RoboticsSciSys)!

It features
* 8 furniture models
* 200+ hours of expert demonstrations
* FurnitureSim: Isaac Gym simulator

🧵👇

1:06