Using a transient thermoreflectance (TTR) technique,several Au films with different thicknesses on glass and SiC substrates are measured for thermal characterization of metallic nano-films,including the electron-phonon coupling factor G,interfacial thermal resistance R,and thermal conductivity K s of the substrate. The rear heating-front detecting (RF) method is used to ensure the femtosecond temporal resolution. An intense laser beam is focused on the rear surface to heat the film,and another weak laser beam is focused on the very spot of the front surface to detect the change in the electron temperature. By varying the optical path delay between the two beams,a complete electron temperature profile can be scanned. Different from the normally used single-layer model,the double-layer model involving interfacial thermal resistance is studied here. The electron temperature cooling profile can be affected by the electron energy transfer into the substrate or the electron-phonon interactions in the metallic films. For multiple-target optimization,the genetic algorithm (GA) is used to obtain both G and R. The experimental result gives a deep understanding of the mechanism of ultra-fast heat transfer in metals. Using a transient thermoreflectance (TTR) technique, several Au films with different thicknesses on glass and SiC substrates are measured for thermal characterization of metallic nano-films, including the electron-phonon coupling factor G, interfacial thermal resistance R, and thermal conductivity Ks of the substrate. The rear heating-front detecting (RF) method is used to ensure the femtosecond temporal resolution. An intense laser beam is focused on the rear surface to heat the film, and another weak laser beam is focused on the very spot of the front surface to detect the change in the electron temperature. different from the normally used single-layer model, the double-layer model involving interfacial thermal resistance is studied here. The electron temperature cooling profile can be affected by the electron energy transfer into the substrate or the electron-phonon interactio ns in the metallic films. For multiple-target optimization, the genetic algorithm (GA) is used to obtain both G and R. The experimental result gives a deep understanding of the mechanism of ultra-fast heat transfer in metals.