Virtual measurement of conditions proximate to a substrate with physics-informed compressed sensing

1. A method, comprising:

measuring a subset of property values in a manufacturing chamber using a plurality of sensors disposed within the manufacturing chamber during a process performed on a substrate in the manufacturing chamber;

determining a map of property values proximate to the substrate in the manufacturing chamber based on inputting measurements of the subset of property values in the manufacturing chamber into a trained machine learning model, wherein the trained machine learning model is configured to represent property values in the manufacturing chamber in accordance with a reduced order model, wherein the map of property values is determined as a weighted additive combination of a number of members of the reduced order model that are a subset of a basis set that describes property values of interest in the manufacturing chamber; and

performing an action in response to the map of property values indicating an abnormality, the action comprising making an adjustment to at least one processing parameter associated with operation of the manufacturing chamber or sending an alert to a client device.

2. The method of claim 1 , wherein the trained machine learning model comprises a sparse regression model, wherein the map of property values is determined using a sparse regression algorithm of the sparse regression model.

3. The method of claim 1 , the basis set having been determined in view of results of a physics-based computer-aided model predicting property values throughout the manufacturing chamber for a plurality of different processing parameter configurations.

4. The method of claim 1 , wherein the map of property values comprises temperatures at a plurality of locations on the substrate.

5. The method of claim 4 , wherein the measuring the subset of property values in the manufacturing chamber comprises taking temperature measurements at a plurality of points, at least a majority of which are not on the substrate, in the manufacturing chamber.

6. A method comprising:

generating training data for a machine learning model, wherein generating the training data comprises:

identifying a first training input comprising data indicative of first property values associated with a limited subset of property values in a manufacturing chamber; and

identifying a first target output for the first training input, wherein the first target output comprises a map of property values in a plurality of locations within the manufacturing chamber, wherein a number of the plurality of locations is more than a number of locations of the subset of property values used as the first training input; and

providing the training data to train the machine learning model on (i) a set of training inputs comprising the first training input, and (ii) a set of target outputs comprising the first target output, wherein the trained machine learning model is to execute on a computing device, and is to receive a new input comprising data indicative of second property values associated with a limited subset of property values in the manufacturing chamber, the second property values being different than the first property values, and to produce a new output based on the new input, the new output comprising a map of property values proximate to a substrate in the manufacturing chamber determined as a weighted additive combination of a number of members of a reduced order model that are a subset of a basis set, and wherein the computing device is to perform at least one of (i) adjusting manufacturing parameters associated with operation of the manufacturing chamber or (ii) sending an alert to a user device, in response to the new output indicating an abnormality.

7. The method of claim 6 , wherein each training input in the set of training inputs is mapped to the target output in the set of target outputs.

8. The method of claim 6 , wherein the training inputs comprise data indicative of a number of measurements taken of property values in the manufacturing chamber at locations not adjacent to the substrate, and the target outputs comprise predictions of property values at locations proximate the substrate.

9. The method of claim 6 , wherein the target outputs comprise data indicative of results of a computer-aided physics-based model, describing property values in the manufacturing chamber for a plurality of processing parameters.

10. The method of claim 9 , wherein a subset of the results of the computer-aided physics-based model, corresponding to locations in the manufacturing chamber where change in the value of a property of interest is above a certain threshold when steady-state conditions with different manufacturing parameters are compared, is used.

11. The method of claim 6 , wherein the machine learning model comprises a sparse regression model.

12. The method of claim 6 , wherein the subset of property values comprises measurements of conditions at a limited number of locations in the manufacturing chamber.

13. The method of claim 12 , wherein the limited number of locations are chosen by inputting results of a computer-aided physics-based model into a second trained machine learning model, an output of the second trained machine learning model comprising optimized locations to take measurements.

14. A non-transitory machine-readable storage medium storing instructions that, when executed by a processing device, cause the processing device to perform operations comprising:

receiving data indicative of measurements of a subset of property values within a manufacturing chamber containing a substrate;

determining, using a trained machine learning model, a map of property values within the manufacturing chamber, comprising property values proximate to the substrate, wherein the trained machine learning model is configured to represent property values of interest in the manufacturing chamber in accordance with a reduced order model, wherein the map of property values is determined as a weighted additive combination of a number of members of the reduced order model that are a subset of a basis set that describes the property values of interest in the manufacturing chamber; and

performing at least one of (i) adjusting manufacturing parameters associated with operation of the manufacturing chamber in view of the map of property values, or (ii) sending an alert to a user device in response to the map of property values indicating an abnormality.

15. The non-transitory machine-readable storage medium of claim 14 , wherein the basis set is constructed in view of results of a computer-aided physics-based model of the manufacturing chamber evaluated at a plurality of processing parameters.

16. The non-transitory machine-readable storage medium of claim 14 , the measurements of the subset of property values in the manufacturing chamber comprising temperature measurements taken at a plurality of points within the manufacturing chamber, at least a majority of which are not on the substrate.

17. The non-transitory machine-readable storage medium of claim 14 , the map of property values comprising temperatures at a plurality of points in the manufacturing chamber, the plurality of points comprising at least points on the substrate.