For modular addition at p=79, 97, and 127, EGD and vanilla SGD follow diverging trajectories and converge to different parameter subspaces. Both overshoot in distance from initialization and later settle in regions having the same fixed distance from initialization, while EGD moves little after its initial large steps and vanilla SGD traverses a wider range. · CiteArk
Not assessedPlan blockedFindingc19-modular-addition-parameter-trajectories
For modular addition at p=79, 97, and 127, EGD and vanilla SGD follow diverging trajectories and converge to different parameter subspaces. Both overshoot in distance from initialization and later settle in regions having the same fixed distance from initialization, while EGD moves little after its initial large steps and vanilla SGD traverses a wider range.
Source: paper:PDF pp. 25-27, Appendix F, Figures 17-18 and accompanying observations
Reported and observed measurements
No structured measurement is attached to this Claim.
Assessments (0)
No immutable Assessment has been published for this Claim yet.
