@rodgerljl Profile picture

Rodger Luo

@rodgerljl

AI Researcher & Artist

Similar User
Ruihan Yang photo

@RchalYang

0xMarcus photo

@Marcus_Rein_

Yifan Wang photo

@toomanyyifans

Carmen Thong photo

@partickler

Vincent photo

@vincentpeaq

Minghao Guo photo

@GuoMh14

GraphBuildersDAO photo

@GraphBuilders

Jeditopher 🇺🇦 | Pinax | Graphtronauts🛡️ photo

@TheRealTopher66

Indexer Office Hours photo

@TheGraphIOH

Nerve photo

@lo4nerve

Rebecca Xu photo

@rebeccaxu_SU

gustavoinacio.eth photo

@GustavoInacio

vincenttaglia.eth photo

@vincenttaglia

jimxjim.eth🦇🔊 photo

@jim74255

Doris Wang photo

@dodo_w929

Rodger Luo Reposted

"Assembly planning is the core of automating product assembly, maintenance, & recycling for modern industrial mfg." Check out the collaboration between @MIT_CSAIL, @ADSKResearch & @TAMU at #SIGGRAPHAsia2022 to learn more! bit.ly/3Vxmkld


Rodger Luo Reposted

Thanks to my wonderful collaborators @xujie7979, Yichen Li, @rodgerljl, Shinjiro Sueda, Hui Li, @KarlDD and my advisor Wojciech Matusik! This work is a collaboration between @MIT_CSAIL, @ADSKResearch and @TAMU


Rodger Luo Reposted

📢JoinABLe is a new #CVPR2022 paper from the @autodesk AI Lab and @MIT_CSAIL Ever wondered how physical objects with hundreds of parts get assembled in CAD? It’s a tedious process that we automate by learning how pair of parts connect to form joints. 1/9🧵

JoinABLe: Learning Bottom-up Assembly of Parametric CAD Joints abs: arxiv.org/abs/2111.12772

Tweet Image 1


Rodger Luo Reposted

I’m super excited to see ideas from complex systems such as swarm intelligence, self-organization, and emergent behavior gain traction again in AI research. We wrote a survey of recent developments that combine ideas from deep learning and complex systems: arxiv.org/abs/2111.14377

Tweet Image 1
Tweet Image 2
Tweet Image 3
Tweet Image 4

Rodger Luo Reposted

This RL agent skipped normal walking and went straight to Michael Jackson. Common robotic RL benchmarks lack an energy penalty, therefore you can get them "solved" (in this case 6000+ pts) with quite energy wasting policies. @hardmaru


Rodger Luo Reposted

Fusion 360 provides a construction sequence of ~9k CAD designs and a gym environment to do CAD reconstruction given a target geometry. arxiv.org/abs/2010.02392

Tweet Image 1

arxiv.org/abs/1910.04376, it’s interesting that it also includes Mahjong.


The goal to use softmax is to make sure one value is very high (close to 1) and all other values are very low (close to 0). Softmax uses exponential to make sure this happens.


Decomposition of Human Portraits 2.0 is featured at ChinaVIS 2019, Chengdu, China. The phots was originally taken by OPPO Rena 10x Zoom, then sent to a neural neural network for visualization. The network also generates a three-word title for each paint.

Tweet Image 1

Rodger Luo Reposted

very elegant visualization of olympic movement by Felix Deimann vimeo.com/100576137


Rodger Luo Reposted

FuelBand Fibers - Visualizing training data with Plask / @variable_io @marcinignac ca-n.in/1lXARi5 | http://t.co/3VfDa2I6hJ

Tweet Image 1

Rodger Luo Reposted

'Learn how to incorporate Kinect v2 into your applications/experiences/installations.' w/ @benlower @KinectWindows ow.ly/Ai2El


Rodger Luo Reposted

Telephone Tower in Stockholm, ca. 1890's - over 5,000 phone lines. Each one a physical cable. http://t.co/c02UU3LiPt

Tweet Image 1

Loading...

Something went wrong.


Something went wrong.