Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for selecting an action to be performed by a reinforcement learning agent. The method includes obtaining an observation characterizing a current state of an environment. For each layer parameter of each noisy layer of a neural network, a respective noise value is determined. For each layer parameter of each noisy layer, a noisy current value for the layer parameter is determined from a current value of the layer parameter, a current value of a corresponding noise parameter, and the noise value. A network input including the observation is processed using the neural network in accordance with the noisy current values to generate a network output for the network input. An action is selected from a set of possible actions to be performed by the agent in response to the observation using the network output.

Abstract: The invention relates to a SERS method for analyzing a biological sample, the method comprising the following step of: a. obtaining a biological sample which is viscous biofluid, b. depositing at least one droplet of the biological sample onto a microscope slide, and drying the droplet, c. depositing a drop of an aqueous dispersion of metallic nanoparticles above the droplet dried in step b), to have a dense distribution of nanoparticles on the surface of the dried droplet and to obtain a SERS-activated biological sample, d. drying the SERS-activated biological sample, e. irradiating the SERS-activated biological sample using a light source to obtain a SERS spectrum, and f. collecting the SERS spectrum.

Abstract: A method for detecting and quantifying fibrosis, which involves at least; positioning a non-colored fixed tissue sample on a support; carrying out infrared acquisition, by scanning, of points of the sample; converting the infrared spectrum produced by each acquired point into at least one spectral image; carrying out a digital processing operation on the spectral image by distinguishing between at least physiological collagen and pathological collagen from fibrosis, attributing, to each point, a spectral class defined by an algorithm classifying the spectra according to the spectral similarities % thereof, then attributing a status to each class by statistically comparing with a digital model of spectra; quantifying the relative surface area occupied by the points of each status; and recording the result.

Abstract: A method for detecting and quantifying fibrosis, which involves at least; positioning a non-coloured fixed tissue sample on a support; carrying out infrared acquisition, by scanning, of points of the sample; converting the infrared spectrum produced by each acquired point into at least one spectral image; carrying out a digital processing operation on the spectral image by distinguishing between at least physiological collagen and pathological collagen from fibrosis, attributing, to each point, a spectral class defined by an algorithm classifying the spectra according to the spectral similarities % thereof, then attributing a status to each class by statistically comparing with a digital model of spectra; quantifying the relative surface area occupied by the points of each status; and recording the result.

Abstract: This invention relates to a method for identifying and classifying carcinomas on the skin of a subject by a FTIR or Raman spectrometer coupled with a micro-imaging system.