HOLISMOKES - XII. Time-delay measurements of strongly lensed Type Ia supernovae using a long short-term memory network

Home

All issues

Volume 692 (December 2024)

A&A, 692 (2024) A132

Full HTML

Open Access

Table 3

Time-delay deviations of the LSTM-FCNN on SNEMO15- only test sets in comparison to the RF and PyCS for a LSN Ia with (κ₁,γ₁, s₁) = (0.25, 0.27, 0.6) for image one and (κ₂,γ₂, s₂) = (0.83, 0.81, 0.6) for image two for z_s = 0.76 and various detection times (in observer days) before the first SN image peaks in brightness.

	LSTM-FCNN	RF	PyCS
At peak detection	$- {0.13}_{- 0.74}^{+ 0.70}$ $- 0.13_{ - 0.74}^{ + 0.70}$	${0.83}_{- 2.05}^{+ 3.33}$ $0.83_{ - 2.05}^{ + 3.33}$	$- {0.16}_{- 3.52}^{+ 3.99}$ $- 0.16_{ - 3.52}^{ + 3.99}$
4 days pre-peak	$- {0.12}_{- 0.57}^{+ 0.54}$ $- 0.12_{ - 0.57}^{ + 0.54}$	${0.22}_{- 1.62}^{+ 2.07}$ $0.22_{ - 1.62}^{ + 2.07}$	${0.12}_{- 2.22}^{+ 2.89}$ $0.12_{ - 2.22}^{ + 2.89}$
6 days pre-peak	$- {0.06}_{- 0.52}^{+ 0.50}$ $- 0.06_{ - 0.52}^{ + 0.50}$	${0.17}_{- 1.48}^{+ 1.61}$ $0.17_{ - 1.48}^{ + 1.61}$	${0.15}_{- 1.57}^{+ 2.25}$ $0.15_{ - 1.57}^{ + 2.25}$
8 days pre-peak	$- {0.05}_{- 0.49}^{+ 0.47}$ ${ - 0.05_{ - 0.49}^{ + 0.47}}$	${0.09}_{- 1.42}^{+ 1.46}$ ${0.09_{ - 1.42}^{ + 1.46}}$	${0.09}_{- 1.39}^{+ 1.87}$ ${0.09_{ - 1.39}^{ + 1.87}}$

Notes. Results for the RF and PyCS are taken from HOLISMOKES VII for SNEMO15 datasets (Huber et al. 2022).

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.