Abstract: The present invention is a system for treating effluent from robotic equipment used to remove marine biofouling, primarily targeting organisms such as orange cup coral, which collect on hulls of floating units for transporting oil and derivatives thereof, or on production and exploration platforms. The system involves separating the flow into two elemental phases: solid and liquid. In the solid-phase flow, treatment is carried out in steps essentially comprising particle-size reduction, direct oxidation, oxidation residence, separation, drying, temporary hermetic storage, packaging, and automatic transportation for final storage. The liquid phase includes a step of centrifuging, adsorption, inerting, and discharge of treated effluent into the sea in accordance with environmental law. The entire system is automated to minimize manual actions and interventions, and is programmed to run without interruption as a function of the flowrate of raw effluent.

Abstract: Persistent storage contains a representation of a neural network including an input layer, and output layer, and a hidden layer, wherein nodes of the hidden layer incorporate serialized activation functions, wherein the serialized activation functions for each of the nodes include a sigmoid function and a Beta function, wherein the sigmoid function is applied to weighted outputs from nodes of a previous layer of the neural network, wherein the Beta function is applied to a conductance hyper-parameter and respective outputs of the sigmoid function, and wherein outputs of the Beta function are provided to a subsequent layer of the neural network. One or more processors are configured to train the neural network until the outputs of the sigmoid function for the nodes of the hidden layer are substantially binary.