splash
67 posts







ok last bug vid before i start capturing bullets with the event sensor. i thought this one was cool because you can see wings flapping near beginning/end of the capture. the reason there are trails is because i'm assigning pixels an exponential decay in post to get this meteor tail effect. this is 10x slower than real time. the event sensor outputs data at 1us timesteps which is nuts.


weekend update: no, noooooooo 😂


What's different between these two BC policies? It's the same architecture, training budget, and data collection setup — the only difference is the controller gains! Controller gains are an understudied design parameter in robot learning. In our new work (w/ @BronarsToni*, @pulkitology), we show how they act as an inductive bias across BC, RL, and Sim2Real transfer, with real consequences on performance. Here's what we found 🧵 * Equal Contribution 📄arxiv: arxiv.org/abs/2604.02523 🔗website: younghyopark.me/tune-to-learn/




یکی از دوستان در حال تولید هستند و به زودی عرضه میشه :)






















