Opened 3 weeks ago

Last modified 37 hours ago

#3014 reviewing enhancement

Extend physics problems to allow testing with simulated noise

Reported by: gkronber Owned by: mkommend
Priority: medium Milestone: HeuristicLab 3.3.16
Component: Problems.Instances Version: trunk
Keywords: Cc:

Description

The physics problem instances for symbolic regression have been introduced with #2957.

It could be helpful to have versions with a small amount of noise.

Attachments (1)

airfoil-lift.png (15.2 KB) - added by gkronber 37 hours ago.

Download all attachments as: .zip

Change History (9)

comment:1 Changed 3 weeks ago by gkronber

  • Status changed from new to accepted
  • Version set to trunk

r17092: changed description of variables and added a copy of the target variable with simulated noise for AircraftLift, FluidDynamics and RocketFuelFlow

comment:2 Changed 3 weeks ago by gkronber

r17093: added a copy of the target variable with simulated noise to the AircraftMaximumLift problem instance

comment:3 Changed 3 weeks ago by gkronber

  • Status changed from accepted to reviewing

comment:4 Changed 3 weeks ago by gkronber

r17094: fixed bugs in AircraftLift problem description

comment:5 Changed 7 days ago by gkronber

  • Owner changed from gkronber to mkommend

comment:6 follow-up: Changed 3 days ago by mkommend

  • Owner changed from mkommend to gkronber
  • Status changed from reviewing to assigned

Reviewed r17092, r17093, r17094.

Changes regarding the noisy target column look good. However, in r17092 the formula for the AircraftLift problem and in r17094 the according description has been changed from alpha - alpha_0 to alpha + alpha_0. In the publication [0] the formula is stated with a minus as it has been originally implemented. Could you please check again which formulation is correct.

[0] https://arxiv.org/pdf/1706.02281.pdf

comment:7 in reply to: ↑ 6 Changed 37 hours ago by gkronber

Replying to mkommend:

[...] However, in r17092 the formula for the AircraftLift problem and in r17094 the according description has been changed from alpha - alpha_0 to alpha + alpha_0. In the publication [0] the formula is stated with a minus as it has been originally implemented. Could you please check again which formulation is correct.

[0] https://arxiv.org/pdf/1706.02281.pdf

Thanks for the critical review. I rechecked the formula in J. Anderson "Fundamentals of Aerodynamics" 5th edition. The zero-lift angle of attack is typically in the range of -2° and -1° and is the angle where the lift of the airfoil is zero. Indeed I introduced an error in r17094. Fixed the formula in r17150.

Changed 37 hours ago by gkronber

comment:8 Changed 37 hours ago by gkronber

  • Owner changed from gkronber to mkommend
  • Status changed from assigned to reviewing
Note: See TracTickets for help on using tickets.