Abstract: A post-deployment drift detection monitoring of a predictive model is described. The method includes accessing a predictive performance metric of a machine learning model that is deployed at a server, the machine learning model being trained with an initial set of training data containing historical data, the predictive performance metric being based on the initial set of training data and an additional set of training data, the additional set of training data containing training data collected since training the machine learning model, detecting a drift based on the predictive performance metric exceeding a drift detection threshold, generating a drift warning notification to a client device, the drift warning notification indicating that the predictive performance metric exceeds the drift detection threshold, receiving a user feedback from the client device, and adjusting one of the machine learning model or the drift detection threshold based on the user feedback.

Abstract: A system for private and secure data portal is described. A method includes receiving, from a first client device, a permission request and a data request for a first dataset that is stored at a second client device, providing the permission request and the data request to the second client device, the second client device configured to generate, in response to the permission request and the data request, a data usage approval document and an encrypted synthesized dataset corresponding to the data usage approval document, the encrypted synthesized dataset includes a synthetic second dataset representative of the first dataset, receiving, from the second client device, the data usage approval document and the encrypted synthesized dataset, performing, at a server, a computation on the encrypted synthesized dataset based on the data request, and providing the data usage approval document and results of the computation to the first client device.

Abstract: A continuous meta-learning system operating at a server is described. In one aspect, a computer-implemented method includes accessing, at the server, a deployed machine learning model, metadata, and new prediction data for the deployed machine learning model, detecting a novel trend in the deployed machine learning model based on the new prediction data, generating label suggestions for the novel trend using metadata, querying a plurality of users to verify the label suggestions, detecting a consensus of the plurality of users, the consensus based on a rate of responses from the plurality of users that confirm the label suggestions, and in response to detecting the consensus, training a new machine learning model based on the new prediction data and the consensus of the plurality of users. The computer-implemented method also includes deploying the new machine learning model at the server.

Abstract: A post-deployment drift detection monitoring of a predictive model is described. The method includes accessing a predictive performance metric of a machine learning model that is deployed at a server, the machine learning model being trained with an initial set of training data containing historical data, the predictive performance metric being based on the initial set of training data and an additional set of training data, the additional set of training data containing training data collected since training the machine learning model, detecting a drift based on the predictive performance metric exceeding a drift detection threshold, generating a drift warning notification to a client device, the drift warning notification indicating that the predictive performance metric exceeds the drift detection threshold, receiving a user feedback from the client device, and adjusting one of the machine learning model or the drift detection threshold based on the user feedback.

Abstract: A system for private and secure data portal is described. A method includes receiving, from a first client device, a permission request and a data request for a first dataset that is stored at a second client device, providing the permission request and the data request to the second client device, the second client device configured to generate, in response to the permission request and the data request, a data usage approval document and an encrypted synthesized dataset corresponding to the data usage approval document, the encrypted synthesized dataset includes a synthetic second dataset representative of the first dataset, receiving, from the second client device, the data usage approval document and the encrypted synthesized dataset, performing, at a server, a computation on the encrypted synthesized dataset based on the data request, and providing the data usage approval document and results of the computation to the first client device.

Abstract: A machine learning platform operating at a server is described. The machine learning platform accesses a dataset from a datastore. A task that identifies a target of a machine learning algorithm from the machine learning platform is defined. The machine learning algorithm forms a machine learning model based on the dataset and the task. The machine learning platform deploys the machine learning model and monitors a performance of the machine learning model after deployment. The machine learning platform updates the machine learning model based on the monitoring.

Abstract: A device (100) for covering the welding area for hot rolling lines consists of an outer container (1), a protective shield (2), a blade-holder (3) with scraping blades, and an actuator (9). Protection occurs by linearly sliding the protective shield (2) outside the outer container (1) up to minimizing the gaps with the billets. The cleaning occurs at the same time as the return to the rest position: the scraping blades, which are integral with the container (1) and are installed in contact with the inner surface of the protective shield (2), slide thereon during the raising, thus colliding with the deposits of solidified material and removing them. Completing the device (100) additionally is a small alternating linear motion of the protective shield (2) during operation to always keep the attachment area of the scraping blades clean, thus avoiding the need for excessive detachment forces at the beginning of the raising.

Abstract: A rolling plant for rolling billets or blooms, the plant defining a longitudinal feeding axis X of said billets or blooms and comprising a feeding area for feeding said billets or blooms; a rolling train; a longitudinal roller table to transport the billets or blooms from said feeding area to said rolling train along said longitudinal feeding axis X; a welding machine adapted to move forwards and backwards along said longitudinal roller table to weld billets or blooms to one another when moving from said feeding area to said rolling train; characterized in that said welding machine is slidingly suspended from a bridge and is adapted to translate parallel to said longitudinal feeding axis X, and said bridge is arranged at the longitudinal roller table and is adapted to translate transversely to said longitudinal feeding axis X, whereby in a first operating position of the bridge the welding machine occupies part of the longitudinal roller table and can move forwards and backwards along said longitudinal fe

Abstract: A device (100) for covering the welding area for hot rolling lines consists of an outer container (1), a protective shield (2), a blade-holder (3) with scraping blades, and an actuator (9). Protection occurs by linearly sliding the protective shield (2) outside the outer container (1) up to minimizing the gaps with the billets. The cleaning occurs at the same time as the return to the rest position: the scraping blades, which are integral with the container (1) and are installed in contact with the inner surface of the protective shield (2), slide thereon during the raising, thus colliding with the deposits of solidified material and removing them. Completing the device (100) additionally is a small alternating linear motion of the protective shield (2) during operation to always keep the attachment area of the scraping blades clean, thus avoiding the need for excessive detachment forces at the beginning of the raising.