Table 4.
Model performance compared with previous works.
| Model | Data | Parameters | Stellar mass [M⊙] | Redshift | Survey | σ(Δfexsitu) | Reference |
|---|---|---|---|---|---|---|---|
| Random forest | TNG100 | ∇ρouter, ∇(g − r)outer, fouterhalo, finnerhalo, Mr, r90 | > 1010.3 | z = 0 − 0.2 | SDSS and HSC | 0.08 | This work |
| Random forest | EAGLE | – | – | – | – | 0.09 | This work |
| Random forest | TNG100 (train), | – | – | – | – | 0.09 | This work |
| EAGLE (validate) | |||||||
| Random forest | EAGLE (train), | – | – | – | – | 0.09 | This work |
| TNG100(validate) | |||||||
| Random forest | TNG100 | M*, Mr, Mg, g − r, r50, r90, C, σ | > 1010.16 | z = 0 | SDSS | ∼0.1 | (Shi et al. 2022) |
| cINN | TNG100 | M*, lookbacktime, Re, fdisk, g − r, Z*, Age* | 1010 − 1012 | z = 0 − 1 | No mock Survey | ∼0.06 | (Eisert et al. 2023) |
| CNN | TNG100 | 2D maps of mass, v, σ, age, metallicity within 1Re | > 1010 | z = 0 | MaNGA | ∼0.07 | (Angeloudi et al. 2023) |
| CNN | EAGLE | – | – | – | – | ∼0.08 | (Angeloudi et al. 2023) |
| CNN | TNG100 (train), | – | – | – | – | ∼0.1 | (Angeloudi et al. 2023) |
| EAGLE(validate) | |||||||
| CNN | EAGLE (train), | – | – | – | – | ∼0.1 | (Angeloudi et al. 2023) |
| TNG100(validate) |
Notes. For this work we only used parameters directly derived from mock photometric images. In comparison, similar works in the literature include parameters or 2D maps of stellar mass, kinematics, age, and metallicity that can only be obtained from spectroscopic or even expensive IFU observations.
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.