Neural Network Data Storage for Capacity and Accuracy Trade-offs
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data storage systems are inefficient in handling large amounts of data generated by modern IT infrastructure, such as smart machines and IoT devices, as they struggle to effectively compress and reconstruct data while maintaining accuracy and storage capacity.
Innovation Solution
A neural network-based data storage system that uses a translation circuit with machine learning circuits to convert original data sets into smaller transformed sets, allowing for lossy or lossless storage, and reconstructing the data when needed, thereby reducing storage requirements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If traditional storage systems store original data sets directly, then data accuracy is maintained, but storage capacity is insufficient for large volumes of data
Solution Approach 1:
The patent transforms data from its original form into a different representation (transformed data set) that occupies less storage space. The neural network learns to map original data to transformed data with reduced dimensions, enabling compact storage while preserving essential information through the learned transformation parameters.
Solution Approach 2:
Instead of storing complete original data sets, the system stores compressed representations (transformed data sets) that capture the essential characteristics. The neural network creates a simplified copy that retains key information while eliminating redundancy, achieving space-efficient storage.
2Productivity
If data is compressed to increase storage capacity, then storage efficiency improves, but reconstruction fidelity deteriorates
Solution Approach 1:
The system uses training data to teach the neural network the relationship between original and transformed data. During training, the network learns to preserve essential features while compressing data. The feedback from training examples enables the network to optimize its transformation to maintain reconstruction quality.
Solution Approach 2:
The neural network is pre-trained on representative data before actual storage operations. This preliminary training phase establishes the optimal transformation parameters that balance compression efficiency with reconstruction fidelity, preparing the system for accurate data recovery.
3Quantity of substance
If neural network transformation is applied to reduce data size, then storage requirements decrease, but system complexity increases
Solution Approach 1:
The neural network automatically learns the optimal transformation from training data without manual intervention. The system self-configures the compression parameters and transformation rules, eliminating the need for complex manual setup and reducing operational complexity despite the advanced technology used.
Data Source
AI summary
According to one general aspect, an apparatus may include a host interface circuit configured to receive a memory access request, wherein the memory access request is associated with a data set. The apparatus may include a non-volatile memory storage circuit configured to create and store a transformed data set. The apparatus may include a translation circuit comprising at least one machine learning circuit. The translation circuit may be configured to: in response to a write memory access, convert an original version of the data set to the transformed data set, and in response to a read memory access, convert the transformed data set to a reconstructed data set.