From 81353f29359243703d5ef8814dc2325fd39dd15b Mon Sep 17 00:00:00 2001 From: rthomasson23 Date: Wed, 17 Apr 2024 14:03:38 -0700 Subject: [PATCH] Update goldberg_24.md --- _talks/goldberg_24.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/_talks/goldberg_24.md b/_talks/goldberg_24.md index 15e2db9..cf7793a 100644 --- a/_talks/goldberg_24.md +++ b/_talks/goldberg_24.md @@ -6,6 +6,6 @@ date: 2024-04-19T12:30:00-0000 location: Skilling Auditorium location-url: "https://campus-map.stanford.edu/?srch=Skillaud" title: "Data is All You Need: Large Robot Action Models and Good Old Fashioned Engineering" -abstract: "2024 is off to an exciting start with enormous enthusiasm for humanoids and other robots based on recent advances in \"end-to-end\" large robot action models. Initial results are promising, and several collaborative efforts are underway to collect the needed demonstration data. But is Data All You Need? I'll present my view of the status quo in terms of manipulation task definition, data collection, and experimental evaluation. I'll then suggest that to reach expected performance levels, it may be helpful for the community to reconsider good old fashioned engineering in terms of modularity, metrics, and failure analysis. I'll present MANIP, a potential framework for doing this that shows promise for tasks such as cable untangling, surgical suturing, and bagging. I welcome feedback: this will be the first time I present this talk and I expect it to be a bit controversial ;)" +abstract: "2024 is off to an exciting start with enormous enthusiasm for humanoids and other robots based on recent advances in \"end-to-end\" large robot action models. Initial results are promising, and several collaborative efforts are underway to collect the needed demonstration data. But is Data All You Need? I'll present my view of the status quo in terms of manipulation task definition, data collection, and experimental evaluation. I'll then suggest that to reach expected performance levels, it may be helpful for the community to reconsider good old fashioned engineering in terms of modularity, metrics, and failure analysis. I'll present MANIP, a potential framework for doing this that shows promise for tasks such as cable untangling, surgical suturing, and bagging. \n I welcome feedback: this will be the first time I present this talk and I expect it to be a bit controversial ;)" youtube-code: "TBD" ---