Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for performing neural architecture search for machine learning models. In one aspect, a method comprises receiving training data for a machine learning, generating a plurality of candidate neural networks for performing the machine learning task, wherein each candidate neural network comprises a plurality of instances of a layer block composed of a plurality of layers, for each candidate neural network, selecting a respective type for each of the plurality of layers from a set of layer types that comprises, training the candidate neural network and evaluating performance scores for the trained candidate neural networks as applied to the machine learning task, and determining a final neural network for performing the machine learning task based at least on the performance scores for the candidate neural networks.

Abstract: A laser annealing device and methods for annealing using the device are disclosed. The laser annealing device includes: a laser light source system (3); a laser adjusting system (4) that is connected to the laser light source system and is disposed above a wafer (1); a temperature monitoring system (5) that is disposed above the wafer (1) and configured to measure in real time a temperature at a location of the surface of the wafer at which a light spot is formed; and a central control system (6) in connection with each of the laser light source system, the laser adjusting system, the temperature monitoring system and a wafer table (2). The wafer is jointly annealed by laser beams from several independent lasers (31) in the laser light source system, which have different wavelengths and cooperate in a mutually complementary manner, with a selected optimum set of process parameters. As a result, an optimum annealing temperature can be achieved and surface pattern effects can be effectively reduced.