International Business Machines Corporation

Globalfoundries Inc.

Jingrui He

Yada Zhu

Robert J Baseman

Emmanuel Yashchin

Richard D Lawrence

Ching-Yung Lin

Enara C Vijil

Yan Liu

Zhen Wen

David C Gondek

Hanghang Tong

Ravi B Konuru

Transfer learning is the task of leveraging the information from labeled examples in some domains to predict the labels for examples in another domain. It finds abundant practical applications, such as sentiment prediction, image classification and network intrusion detection. A graph-based transfer learning framework propagates label information from a source domain to a target domain via the example-feature-example tripartite graph, and puts more emphasis on the labeled examples from the target domain via the example-example bipartite graph. An iterative algorithm renders the framework scalable to large-scale applications. The framework propagates the label information to both features irrelevant to the source domain and unlabeled examples in the target domain via common features in a principled way.

Graph-based transfer learning

The present invention generally relates to the monitoring and controlling of a semiconductor manufacturing environment and, more particularly, to methods and systems for virtual meteorology (VM) applications based on data from multiple tools having heterogeneous relatedness. The methods and systems leverage the natural relationship of the multiple tools and take advantage of the relationship embedded in process variables to improve the prediction performance of the VM predictive wafer quality modeling. The prediction results of the methods and systems can be used as a substitute for or in conjunction with actual metrology samples in order to monitor and control a semiconductor manufacturing environment, and thus reduce delays and costs associated with obtaining actual physical measurements.

System for wafer quality predictive modeling based on multi-source information with heterogeneous relatedness

An apparatus for performing run-to-run control and sampling optimization in a semiconductor manufacturing process includes at least one control module. The control module is operative: to determine a process output and corresponding metrology error associated with an actual metrology for a current processing run in the semiconductor manufacturing process; to determine a predicted process output and corresponding prediction error associated with a virtual metrology for the current processing run; and to control at least one parameter corresponding to a subsequent processing run as a function of the metrology error and the prediction error.

Run-to-run control utilizing virtual metrology in semiconductor manufacturing

A method for run-to-run control and sampling optimization in a semiconductor manufacturing process includes the steps of: determining a process output and corresponding metrology error associated with an actual metrology for a current processing run in the semiconductor manufacturing process; determining a predicted process output and corresponding prediction error associated with a virtual metrology for the current processing run; and controlling at least one parameter corresponding to a subsequent processing run as a function of the metrology error and the prediction error.

Method and system for wafer quality predictive modeling based on multi-source information with heterogeneous relatedness

A method, system and computer program product for finding a diversified ranking list for a given query. In one embodiment, a multitude of date items responsive to the query are identified, a marginal score is established for each data item; and a set, or ranking list, of the data items is formed based on these scores. This ranking list is formed by forming an initial set, and one or more data items are added to the ranking list based on the marginal scores of the data items. In one embodiment, each of the data items has a measured relevance and a measured diversity value, and the marginal scores for the data items are based on the measured relevance and the measured diversity values of the data items.

Finding a top-K diversified ranking list on graphs

A system and method a Multi-Task Multi-View (MTV) learning problem. The method uses the label information from related tasks to make up for the lack of labeled data in a single task. The method further uses the consistency among different views to improve the performance. It is tailored for the above complicated dual heterogeneous problems where multiple related tasks have both shared and task-specific views (features), since it makes full use of the available information.

Graph-based framework for multi-task multi-view learning

A method for performing enhanced wafer quality prediction in a semiconductor manufacturing process includes the steps of: obtaining data including at least one of tensor format wafer processing conditions, historical wafer quality measurements and prior knowledge relating to at least one of the semiconductor manufacturing process and wafer quality; building a hierarchical prediction model including at least the tensor format wafer processing conditions; and predicting wafer quality for a newly fabricated wafer based at least on the hierarchical prediction model and corresponding tensor format wafer processing conditions.

Growing community of inventors

Yorktown Heights, NY, United States of America

Jingrui He