Abstract: In a method for training a federated learning model, a server obtains a target split mode corresponding to a training node in response to determining that the training node satisfies a preset splitting condition. The server notifies a client to perform, based on the target split mode, node splitting. The server performs a next round of training by taking a left subtree node generated by performing the node splitting as a new training node until an updated training node does not satisfy the preset splitting condition. The server performs a next round of training by taking another non-leaf node of the boosting tree as a new training node. The server stops training and generates a target federated learning model in response to determining that a node dataset of the plurality of boosting trees is empty.

Abstract: Lumber sorting systems and methods that use adaptive lumber management to tailor the distribution of lumber to improve the yield and aesthetic properties of a product are disclosed. An adaptive lumber management system in accordance with the present disclosure may comprise a lumber scanner configured to record one or more characteristics of each board and create a data file for each board comprising information regarding the characteristic of the board. A processor may be configured to determine a target tray for each board based on the data files. The processor may be further configured to generate at least one mapping. The system may further comprise a sorting mechanism configured to sort the plurality of boards into their respective target trays and at least one tipple configured to select each board or the plurality of boards from the plurality of trays according to the at least one mapping.