International Business Machines Corporation

The Boeing Company

Nam H Nguyen

Jayant R Kalagnanam

Ramachandran Muralidhar

Young Min Lee

Robert J Baseman

Christian Makaya

Richard Joel Thompson

Kyong Min Yeo

Edward John Pring

Marna M Kagele

Hui Qi

Dung Tien Phan

John Rubrecht Lowell

Dzung Tien Phan

Kristen Smith Williams

Fateh Ali Tipu

Dhaval Patel

Anuradha Bhamidipaty

Julia Elizabeth Rice

Wesley M Gifford

Lam Nguyen

Dhavalkumar C Patel

Fateh A Tipu

Bhanukiran Vinzamuri

Hsien-Lin Huang

Mario Motta

Tanvi Pradeep Gujarati

Jason Robert Lathrop

Tyler Benjamin Smith

Wilbur Harrison Barwick

Systems, computer-implemented methods, and/or computer program products to facilitate reduction of a quantum circuit are provided. A computer-implemented method can comprise performing, by a system operatively coupled to at least one processor, decomposition of an exponential of a first Pauli operator of 1 to n Pauli operators of the quantum circuit, and performing, by the system, a first Clifford transformation of a primary operator of the quantum circuit, where the primary operator can comprise a linear combination of primary Pauli operators, and where the first Clifford transformation can employ a result of the decomposition. Performing the decomposition can comprise decomposing the exponential of the first Pauli operator into a first non-Clifford operator and a first Clifford operator, with the first Clifford operator being employed for the first Clifford transformation. Additional decompositions and respective transformations can be performed iteratively until decomposition of all exponentials of the 1 to n Pauli operators.

Circuit reduction for exponentials of Pauli operators

Arranging computational sub-tasks in a hybrid-computing environment is provided. The method comprises receiving input of a number of nodes, wherein each node represents a computational sub-task, and wherein the nodes are grouped into different sets according to differing computing resources used by the nodes. A computational objective is also received as well as initial data inputs and desired final outputs. A directed graph network is generated comprising the nodes and directed edges connecting the nodes. An optimization problem is then solved to determine a best path through the directed graph network for deriving the desired final outputs from the initial data inputs according to the computational objective. The best path comprises a subset of nodes and directed edges within the directed graph network. The best path to achieve the computational objective and estimated computational resources required for the best path are output to a user.

Hybrid-computing resource optimization

A method, apparatus, computer system, and computer program product for parallelizing quantum processes for processing a problem. A computer system identifies subproblems in the problem based on a structure of the problem. The computer system identifies quantum circuits in a set of quantum computers to process the subproblems. The computer system executes jobs on the quantum circuits to solve the subproblems.

Quantum process parallelization

A method of minimizing a cost function of a quantum computation is provided. The method comprises receiving input of an initial state of a quantum problem instance comprising a Hamiltonian with an associated cost function. The Hamiltonian is converted into a number of Pauli strings, which are used to form an operator pool. The Pauli strings in the operator pool are ranked according to how much they lower a value of the cost function with respect to the initial state. Pauli strings are iteratively added from the operator pool to a parameterized quantum circuit, in a manner to minimize circuit depth, until a variational quantum eigensolver (VQE) algorithm converges to an approximate ground state wave function generated by the parameterized quantum circuit.

Reducing resources in quantum circuits

A method for process control using predictive long short term memory includes obtaining historical post-process measurements taken on prior products of the manufacturing process; obtaining historical in-process measurements taken on prior workpieces during the manufacturing process; training a neural network to predict each of the historical post-process measurements, in response to the corresponding historical in-process measurements and preceding historical post-process measurements; obtaining present in-process measurements on a present workpiece during the manufacturing process; predicting a future post-process measurement for the present workpiece, by providing the present in-process measurements and the historical post-process measurements as inputs to the neural network; and adjusting at least one controllable variable of the manufacturing process in response to the prediction of the future post-process measurement.

Tool control using multistage LSTM for predicting on-wafer measurements

In an approach for compressing a neural network, a processor receives a neural network, wherein the neural network has been trained on a set of training data. A processor receives a compression ratio. A processor compresses the neural network based on the compression ratio using an optimization model to solve for sparse weights. A processor re-trains the compressed neural network with the sparse weights. A processor outputs the re-trained neural network.

Compression of deep neural networks

A method of improving at least one of quality and yield of a physical process comprises: obtaining values, from respective performances of the physical process, for a plurality of variables associated with the physical process; determining at least one Gaussian mixture model (GMM) representing the values for the variables for the performances of the physical process; based at least in part on the at least one GMM, computing at least one anomaly score for at least one of the variables for at least one of the performances of the physical process; based on the at least one anomaly score, identifying the at least one of the performances of the physical process as an outlier; and, based at least in part on the outlier identification, modifying the at least one of the variables for one or more subsequent performances of the physical process.

Model management for non-stationary systems

A computer implemented method of administering a complex system includes receiving multivariate data from a plurality of sensors of the system in an ambient state. Event sequences in the received multivariate data are identified. The multivariate event sequences are projected to a lower stochastic latent embedding. A temporal structure of the sequences is learned in a lower latent space. A probabilistic prediction in the lower latent space is provided. The probabilistic prediction in the lower stochastic latent space is decoded to an event prediction in the ambient state.

Event time characterization and prediction in multivariate event sequence domains to support improved process reliability

Anomaly detection and remedial recommendation techniques for improving the quality and yield of microelectronic products are provided. In one aspect, a method for quality and yield improvement via anomaly detection includes: collecting time series sensor data during individual steps of a semiconductor manufacturing process; calculating anomaly scores for each of the individual steps using a predictive model; and implementing changes to the semiconductor manufacturing process based on the anomaly scores. A system for quality and yield improvement via anomaly detection is also provided.

Anomaly detection and remedial recommendation

Based on at least one manufacturing process characteristics associated with a manufacturing process, a prediction time at which to execute a selected machine learning model selected from multiple trained machine learning models is determined, and at the prediction time, the selected machine learning model is executed. Executing the selected machine learning model predicts a control set point for future values of state variables of the manufacturing process, for controlling the manufacturing process. Based on at least one of the manufacturing process characteristics, a learning time at which to train a machine learning model is determined, and at the learning time, the machine learning model is trained based on historical process data associated with the manufacturing process.

Orchestration of learning and execution of model predictive control tool for manufacturing processes

Methods and systems for determining an optimized set point for a manufacturing apparatus are described. In an example, a processor may receive observed data from the manufacturing apparatus. The observed data may include data collected by the manufacturing apparatus based on at least one resolution. The processor may generate feature data based on the observed data. The processor may determine a first model and a second model based on the feature data. The first model may relate to a first prediction of a key performance indicator of the manufacturing apparatus in a first amount of future time. The second model may relate to a second prediction of the key performance indicator of the manufacturing apparatus in a second amount of future time. The processor may determine the optimized set point based on an objective relating to the first model and based on a constraint relating to the second model.

Growing community of inventors

Pleasantville, NY, United States of America

Nam H Nguyen