Zobrazeno 1 - 10
of 166
pro vyhledávání: '"Gross, Jason P."'
A recent line of work in mechanistic interpretability has focused on reverse-engineering the computation performed by neural networks trained on the binary operation of finite groups. We investigate the internals of one-hidden-layer neural networks t
Externí odkaz:
http://arxiv.org/abs/2410.07476
Natural environments pose significant challenges for autonomous robot navigation, particularly due to their unstructured and ever-changing nature. Hiking trails, with their dynamic conditions influenced by weather, vegetation, and human traffic, repr
Externí odkaz:
http://arxiv.org/abs/2409.15671
Autor:
Gross, Jason, Agrawal, Rajashree, Kwa, Thomas, Ong, Euan, Yip, Chun Hei, Gibson, Alex, Noubir, Soufiane, Chan, Lawrence
We propose using mechanistic interpretability -- techniques for reverse engineering model weights into human-interpretable algorithms -- to derive and compactly prove formal guarantees on model performance. We prototype this approach by formally prov
Externí odkaz:
http://arxiv.org/abs/2406.11779
Autor:
Smith, Trevor, Rijal, Madhav, Tatsch, Christopher, Butts, R. Michael, Beard, Jared, Cook, R. Tyler, Chu, Andy, Gross, Jason, Gu, Yu
This work presents the design of Stickbug, a six-armed, multi-agent, precision pollination robot that combines the accuracy of single-agent systems with swarm parallelization in greenhouses. Precision pollination robots have often been proposed to of
Externí odkaz:
http://arxiv.org/abs/2404.03489
Publikováno v:
NAVIGATION: Journal of the Institute of Navigation December 2023, 70 (4) navi.608
This paper proposes the cooperative use of zero velocity update (ZU) in a decentralized extended Kalman filter (DEKF) based localization algorithm for multi-robot systems. The filter utilizes inertial measurement unit (IMU), ultra-wideband (UWB), and
Externí odkaz:
http://arxiv.org/abs/2306.17703
Autor:
Kuepper, Joel, Erbsen, Andres, Gross, Jason, Conoly, Owen, Sun, Chuyue, Tian, Samuel, Wu, David, Chlipala, Adam, Chuengsatiansup, Chitchanok, Genkin, Daniel, Wagner, Markus, Yarom, Yuval
Manual engineering of high-performance implementations typically consumes many resources and requires in-depth knowledge of the hardware. Compilers try to address these problems; however, they are limited by design in what they can do. To address thi
Externí odkaz:
http://arxiv.org/abs/2305.19586
Publikováno v:
J Autom Reasoning 68, 19 (2024)
We address the challenges of scaling verification efforts to match the increasing complexity and size of systems. We propose a research agenda aimed at building a performant proof engine by studying the asymptotic performance of proof engines and red
Externí odkaz:
http://arxiv.org/abs/2305.02521
Autor:
Kuepper, Joel, Erbsen, Andres, Gross, Jason, Conoly, Owen, Sun, Chuyue, Tian, Samuel, Wu, David, Chlipala, Adam, Chuengsatiansup, Chitchanok, Genkin, Daniel, Wagner, Markus, Yarom, Yuval
Most software domains rely on compilers to translate high-level code to multiple different machine languages, with performance not too much worse than what developers would have the patience to write directly in assembly language. However, cryptograp
Externí odkaz:
http://arxiv.org/abs/2211.10665
Slip detection is of fundamental importance for the safety and efficiency of rovers driving on the surface of extraterrestrial bodies. Current planetary rover slip detection systems rely on visual perception on the assumption that sufficient visual f
Externí odkaz:
http://arxiv.org/abs/2207.13629
Autor:
Das, Shounak, Gross, Jason
In this article, we present a method for increasing adaptivity of an existing robust estimation algorithm by learning two parameters to better fit the residual distribution. The analyzed method uses these two parameters to calculate weights for Itera
Externí odkaz:
http://arxiv.org/abs/2206.10305