17971410. DATA-AWARE STORAGE TIERING AND LIFETIME DATA VALUATION FOR DEEP LEARNING simplified abstract (HEWLETT PACKARD ENTERPRISE DEVELOPMENT LP)
Contents
- 1 DATA-AWARE STORAGE TIERING AND LIFETIME DATA VALUATION FOR DEEP LEARNING
- 1.1 Organization Name
- 1.2 Inventor(s)
- 1.3 DATA-AWARE STORAGE TIERING AND LIFETIME DATA VALUATION FOR DEEP LEARNING - A simplified explanation of the abstract
- 1.4 Simplified Explanation
- 1.5 Key Features and Innovation
- 1.6 Potential Applications
- 1.7 Problems Solved
- 1.8 Benefits
- 1.9 Commercial Applications
- 1.10 Prior Art
- 1.11 Frequently Updated Research
- 1.12 Questions about Machine Learning Model Training
- 1.13 Original Abstract Submitted
DATA-AWARE STORAGE TIERING AND LIFETIME DATA VALUATION FOR DEEP LEARNING
Organization Name
HEWLETT PACKARD ENTERPRISE DEVELOPMENT LP
Inventor(s)
SUPARNA Bhattacharya of Bangalore (IN)
RYAN Beethe of Houston TX (US)
MARTIN Foltin of Ft. Collins CO (US)
DATA-AWARE STORAGE TIERING AND LIFETIME DATA VALUATION FOR DEEP LEARNING - A simplified explanation of the abstract
This abstract first appeared for US patent application 17971410 titled 'DATA-AWARE STORAGE TIERING AND LIFETIME DATA VALUATION FOR DEEP LEARNING
Simplified Explanation
The patent application describes systems and methods for providing lifetime data valuations for a dataset that evolves across multiple machine learning training tasks by updating path-dependent data valuations for data points in the dataset during each training task.
- The system splits the dataset into random mini-epochs for training the machine learning model.
- It uses high-value data points from accuracy mini-epochs to train the model effectively.
- The path-dependent data valuations are updated based on the similarity between the current and prior trained machine learning models.
Key Features and Innovation
- Path-dependent data valuations for evolving datasets.
- Training using random and accuracy mini-epochs.
- Updating data valuations based on model similarity.
Potential Applications
This technology can be applied in various fields such as finance, healthcare, and marketing for improving machine learning model training and accuracy.
Problems Solved
This technology addresses the challenge of adapting machine learning models to evolving datasets and improving the accuracy of training tasks.
Benefits
- Enhanced accuracy in machine learning training.
- Improved adaptability to evolving datasets.
- Efficient updating of data valuations.
Commercial Applications
Title: Enhanced Machine Learning Model Training for Evolving Datasets This technology can be used in industries such as finance for predicting market trends, in healthcare for diagnosing diseases, and in marketing for customer segmentation.
Prior Art
Readers can explore prior research on machine learning model training techniques and data valuation methods to understand the background of this technology.
Frequently Updated Research
Researchers are constantly developing new methods for improving machine learning model training on evolving datasets. Stay updated on the latest advancements in this field.
Questions about Machine Learning Model Training
How does this technology improve the accuracy of machine learning models on evolving datasets?
This technology enhances accuracy by updating data valuations based on the similarity between current and prior trained models, ensuring better adaptation to evolving data.
What are the potential applications of this technology beyond machine learning model training?
This technology can be applied in various industries such as finance, healthcare, and marketing for tasks like predicting market trends, diagnosing diseases, and customer segmentation.
Original Abstract Submitted
Systems and methods are configured to provide lifetime data valuations for a dataset that evolves across multiple machine learning training tasks by providing and updating path-dependent data valuations for data points in the dataset during each training task. A current machine learning training task may include splitting the dataset into multiple random mini-epochs and training the current machine learning model using a first random mini-epoch and an accuracy mini-epoch, which consists of high value data points from the path-dependent data valuations. The random and accuracy mini-epochs can be, during the training, iterated for a number of times during the training, while a second random mini-epoch is prefetch. During the training, the path-dependent data valuations can be updated based on data valuations during the current training and a similarity between the current machine learning model and prior trained machine learning models.
