The example systems, methods, and/or computer-readable media described herein help with design of highly accurate models of un-erupted or partially erupted teeth and help fabricate of aligners for un-erupted or partially erupted teeth. Automated agents that use machine learning models to parametrically represent three-dimensional (3d) virtual representations of teeth as 3D descriptors in a 3D descriptor space are provided herein. In some implementations, the automated agents described herein provide instructions to fabricate aligners for at least partially un-erupted teeth using representative 3D descriptor(s) of a tooth type.

The program code, when executed by a processor, causes the processor to input fabrication data including a plurality of parameters associated with a semiconductor fabricating process to a framework to generate a first class for analyzing the fabrication data, to extract a first parameter targeted for analysis and a second parameter associated with the first parameter from the plurality of parameters and generate a second class for analyzing the first parameter as a sub class of the first class, to modify the first parameter and the second parameter into a data structure having a format appropriate to store in the second class, so as to be stored in the second class, to perform data analysis on the first parameter and the second parameter, to transform the first parameter and the second parameter into corresponding tensor data, and to input the tensor data to the machine learning model.

A method of monitoring a milling machine includes deploying an untrained machine learning model for determining one or more anomalies in time series data. During operation of the milling machine, first time series data representing a rotational speed of a milling head of the milling machine and at least one further operating parameter of the milling machine are obtained by the untrained machine learning model. The untrained machine learning model is trained, during operation of the milling machine, based on the obtained first time series data. Second time series data representing the rotational speed of the milling head of the milling machine and the further operating parameter are obtained by the trained machine learning model during operation of the milling machine. One or more anomalies in the second time series data are determined by the trained machine learning model during operation of the milling machine.

A state determination device that determines an operation state of an injection molding machine stores respective specification data of a reference injection molding machine and an injection molding machine that is different from the reference injection molding machine, and acquires data related to the injection molding machine. Then, the state determination device converts the acquired data into yardstick data by a conversion formula set for every type of data, by using the stored specification data of the reference injection molding machine and the stored specification data of the injection molding machine and performs machine learning using the yardstick data obtained through the conversion so as to generate a learning model.

A computer-implemented method of providing a machine learning model for condition monitoring of an electric power converter is provided. The method includes: obtaining a first batch of input data that includes a number of samples of one or more operating parameters of the converter during at least one operating state of the converter; reducing the number of samples of the first batch by clustering the samples of the first batch into a first set of clusters, (e.g., according to a first clustering algorithm, e.g., based on a clustering feature tree, such as BIRCH), and determining at least one representative sample for each cluster; providing the representative samples for training the machine learning model; and/or training the machine learning model based on the representative samples.

A machine tool includes: a spindle that causes a tool to rotate and move; a workpiece rotation mechanism that causes a workpiece W to rotate; a control unit that controls the spindle and the workpiece rotation mechanism in accordance with commands from a program; and a cutting load detection unit that detects a cutting load imparted on the workpiece by the tool, and the control unit controls a cutting route such that a cutting depth of the workpiece cut with the tool in a region with a small cutting load is greater than the cutting depth in a region with a large cutting load within such a range that the cutting load detected by the cutting load detection unit does not exceed a predetermined load.

Provided are systems and methods for training a robot. The method commences with collecting, by the robot, sensor data from a plurality of sensors of the robot. The sensor data may be related to a task being performed by the robot based on an artificial intelligence (AI) model. The method may further include determining, based on the sensor data and the AI model, that a probability of completing the task is below a threshold. The method may continue with sending a request for operator assistance to a remote computing device and receiving, in response to sending the request, teleoperation data from the remote computing device. The method may further include causing the robot to execute the task based on the teleoperation data. The method may continue with generating training data based on the sensor data and results of execution of the task for updating the AI model.

Implementations utilize deep reinforcement learning to train a policy neural network that parameterizes a policy for determining a robotic action based on a current state. Some of those implementations collect experience data from multiple robots that operate simultaneously. Each robot generates instances of experience data during iterative performance of episodes that are each explorations of performing a task, and that are each guided based on the policy network and the current policy parameters for the policy network during the episode. The collected experience data is generated during the episodes and is used to train the policy network by iteratively updating policy parameters of the policy network based on a batch of collected experience data. Further, prior to performance of each of a plurality of episodes performed by the robots, the current updated policy parameters can be provided (or retrieved) for utilization in performance of the episode.

Provided is a thermal displacement correction method using a machine learning method but making it possible to, on a user side, calculate a thermal displacement amount appropriate to a machine tool of the user and correct the thermal displacement. In a machine tool on a target user side, a thermal displacement amount between workpiece and tool corresponding to a temperature at a preset measurement point is calculated based on a parameter defining a relation between the temperature and the thermal displacement amount, and a positioning position for workpiece and tool is corrected in accordance with the calculated thermal displacement amount. On a manufacturer side, operational status information of the machine tool on the target user side is obtained, an operational status identical to the obtained operational status on the target user side is reproduced with a machine tool of a same type as the machine tool on the target user side based on the obtained operational status information, a temperature at a measurement point identical to the measurement point on the machine tool on the target user side and a thermal displacement amount between workpiece and tool are measured during reproduction, and the parameter is calculated by machine learning based on the measured temperature and thermal displacement amount. The parameter in the machine tool on the target user side is updated with the calculated parameter.

A method includes receiving, by a processing device, first data. The first data includes data from one or more sensors of a processing chamber and is associated with a processing operation. The first data is resolved in at least two dimensions, one of which is time. The method further includes providing the first data to a model. The method further includes receiving from the model second data. The second data includes an indication of an evolution of a processing parameter during the processing operation. The method further includes causing performance of a corrective action in view of the second data.