# US 12,762,379 B2 · Provisioning deep learning (DL) models that preserve relationships between response variables and selected explanatory variables > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,762,379 B2 - Document ID: US12762379B2 - Status: Granted - Kind: B2 - Application number: 18/334,697 - Filed: 2023-06-14 - Published: 2024-09-12 - Granted: 2026-10-13 - Earliest claimed: 2023-03-07 - Assignees: Accenture Global Solutions Limited - Applicants: Accenture Global Solutions Limited - Inventors: Georgios Passalis; Neha Chopra; Sneha Chate; Boobesh Rajendran; Shiva Tyagi; Rajan Gaba - Web page: https://patentgrep.com/patent/US12762379B2 ## CPC - G06N 3/0455: Auto-encoder networks; Encoder-decoder networks - G06N 3/088: Non-supervised learning, e.g. competitive learning - G06N 3/09: Supervised learning - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract Implementations for training a denoising stacked autoencoder (DAE) using a noisy training dataset comprising a noisy sub-set and a non-noisy sub-set, providing an artificial neural network (ANN) including multiple hidden layers, at least one hidden layer including at least a portion of an encoder of the DAE, the at least a portion of the encoder comprising parameters determined during training of the DAE, training the ANN using a training dataset, and providing a version of the ANN for inference. ## Claims ### Claim 1 A computer-implemented method for training deep learning (DL) models, the method comprising: training a denoising stacked autoencoder (DAE) using a noisy training dataset comprising a noisy sub-set and a non-noisy sub-set; wherein the noisy sub-set is generated by: selecting a pre-defined percentage of training samples from a training dataset and copying the selected training samples into the noisy sub-set, and randomly adjusting data attributes of the selected training samples across rows to introduce noise; providing an artificial neural network (ANN) comprising multiple hidden layers, wherein at least one hidden layer comprising at least a portion of an encoder of the trained DAE, and wherein the at least a portion of the encoder comprising parameters of the at least one hidden layer determined during training of the DAE; training the ANN using the training dataset; and providing a version of the ANN for inference. ### Claim 2 (depends on 1) The method of claim 1, wherein training of the DAE comprises unsupervised training. ### Claim 3 (depends on 1) The method of claim 1, wherein training of the ANN comprises supervised training. ### Claim 4 (depends on 1) The method of claim 1, wherein the ANN comprises a classification model that is trained to predict a class in a set of classes. ### Claim 5 (depends on 1) The method of claim 1, further comprising tuning the ANN to provide multiple versions of the ANN, the version of the ANN provided for inference determined to be a best performing version of the multiple versions. ### Claim 6 (depends on 5) The method of claim 5, wherein the ANN is tuned based on one or more of activation function, learning rate, number of neurons, optimizer, batch size, and number of epochs. ### Claim 7 (depends on 1) The method of claim 1, further comprising adding a drop-out layer to the DAE, wherein the drop-out layer deactivates a pre-defined percentage of activations in at least one of hidden layer provided as input to the drop-out layer during training. ### Claim 8 A system, comprising: one or more processors; and a computer-readable storage device coupled to the one or more processors and having instructions stored thereon which, when executed by the one or more processors, cause the one or more processors to perform operations for training deep learning (DL) models, the operations comprising: training a denoising stacked autoencoder (DAE) using a noisy training dataset comprising a noisy sub-set and a non-noisy sub-set; wherein the noisy sub-set is generated by: selecting a pre-defined percentage of training samples from a training dataset and copying the selected training samples into the noisy sub-set, and randomly adjusting data attributes of the selected training samples across rows to introduce noise; providing an artificial neural network (ANN) comprising multiple hidden layers, wherein at least one hidden layer comprising at least a portion of an encoder of the trained DAE, and wherein the at least a portion of the encoder comprising parameters of the at least one hidden layer determined during training of the DAE; training the ANN using the training dataset; and providing a version of the ANN for inference. ### Claim 9 (depends on 8) The system of claim 8, wherein training of the DAE comprises unsupervised training. ### Claim 10 (depends on 8) The system of claim 8, wherein training of the ANN comprises supervised training. ### Claim 11 (depends on 8) The system of claim 8, wherein the ANN comprises a classification model that is trained to predict a class in a set of classes. ### Claim 12 (depends on 8) The system of claim 8, wherein operations further comprise tuning the ANN to provide multiple versions of the ANN, the version of the ANN provided for inference determined to be a best performing version of the multiple versions. ### Claim 13 (depends on 12) The system of claim 12, wherein the ANN is tuned based on one or more of activation function, learning rate, number of neurons, optimizer, batch size, and number of epochs. ### Claim 14 (depends on 8) The system of claim 8, further comprising adding a drop-out layer to the DAE, wherein the drop-out layer deactivates a pre-defined percentage of activations in at least one of hidden layer provided as input to the drop-out layer during training. ### Claim 15 Computer-readable storage media coupled to the one or more processors and having instructions stored thereon which, when executed by the one or more processors, cause the one or more processors to perform operations for training deep learning (DL) models, the operations comprising: training a denoising stacked autoencoder (DAE) using a noisy training dataset comprising a noisy sub-set and a non-noisy sub-set; wherein the noisy sub-set is generated by: selecting a pre-defined percentage of training samples from a training dataset and copying the selected training samples into the noisy sub-set, and randomly adjusting data attributes of the selected training samples across rows to introduce noise; providing an artificial neural network (ANN) comprising multiple hidden layers, wherein at least one hidden layer comprising at least a portion of an encoder of the trained DAE, and wherein the at least a portion of the encoder comprising parameters of the at least one hidden layer determined during training of the DAE; training the ANN using the training dataset; and providing a version of the ANN for inference. ### Claim 16 (depends on 15) The computer-readable storage media of claim 15, wherein training of the DAE comprises unsupervised training. ### Claim 17 (depends on 15) The computer-readable storage media of claim 15, wherein training of the ANN comprises supervised training. ### Claim 18 (depends on 15) The computer-readable storage media of claim 15, wherein the ANN comprises a classification model that is trained to predict a class in a set of classes. ### Claim 19 (depends on 15) The computer-readable storage media of claim 15, wherein operations further comprise tuning the ANN to provide multiple versions of the ANN, the version of the ANN provided for inference determined to be a best performing version of the multiple versions. ### Claim 20 (depends on 15) The computer-readable storage media of claim 15, further comprising adding a drop-out layer to the DAE, wherein the drop-out layer deactivates a pre-defined percentage of activations in at least of hidden layer provided as input to the drop-out layer during training.