notesum.ai

Published at December 3

Recovering implicit physics model under real-world constraints

cs.LG

cs.AI

Released Date: December 3, 2024

Authors: Ayan Banerjee¹, Sandeep K. S. Gupta

Aff.: ¹IMPACT Lab, Arizona State University

Arxiv: http://arxiv.org/pdf/2412.02215v1

Refer to caption

Example	Variables	Inputs	Implicit	Uncertain timing	Nyquist rate	Max sampling rate	No of coefficients
Real-World: Lotka Volterra (B)	$x_{1}$ , $x_{2}$	1	$x_{1}$	No	2.5 Hz	10 Hz	4
Simulation: Chaotic Lorenz System (B)	$x_{1}$ , $x_{2}$ , $x_{3}$	1	$x_{1}$ $x_{2}$	No	100Hz	1000 Hz	4
Simulation: F8 Crusader tracking (B)	$x_{1}$ , $x_{2}$ , $x_{3}$	1	$x_{2}$	No	100 Hz	1000 Hz	20
Simulation: Pathogenic attack (B)	$x_{1}$ , $x_{2}$ , $x_{3}$ $x_{4}$ , $x_{5}$	1	$x_{3}$ , $x_{4}$	No	2.8 $\times$ $10^{-4}$	5.6 $\times$ $10^{-4}$	13
Real-world: Automated Insulin Delivery (N)	$I_{s}$ $G$ $I$	2	$I_{s}$ , $I$	Yes	0.0028 Hz	0.0033 Hz	9
Simulation: AID (N)	$I_{s}$ $G$ $I$	2	$I_{s}$ , $I$	Yes	0.0028 Hz	10 Hz	9
Real -world: EEG (N)	$x_{1}$ , $x_{2}$ , $\dot{x}_{1}$ , $\dot{x}_{2}$	1	$x_{1}$ , $\dot{x_{1}}$	Yes	250 Hz	500 Hz	6