View source for 18574903. ADAPTIVE BUFFER MANAGEMENT TO SUPPORT DYNAMIC TENSOR SHAPE IN DEEP NEURAL NETWORK APPLICATIONS simplified abstract (Intel Corporation)

=ADAPTIVE BUFFER MANAGEMENT TO SUPPORT DYNAMIC TENSOR SHAPE IN DEEP NEURAL NETWORK APPLICATIONS=


==Organization Name==

[[:Category:Intel Corporation|Intel Corporation]]


[[Category:Intel Corporation]]

==Inventor(s)==

[[:Category:Liyang Ling of Shanghai (CN)|Liyang Ling of Shanghai (CN)]][[Category:Liyang Ling of Shanghai (CN)]]

==ADAPTIVE BUFFER MANAGEMENT TO SUPPORT DYNAMIC TENSOR SHAPE IN DEEP NEURAL NETWORK APPLICATIONS - A simplified explanation of the abstract==


This abstract first appeared for US patent application 18574903 titled 'ADAPTIVE BUFFER MANAGEMENT TO SUPPORT DYNAMIC TENSOR SHAPE IN DEEP NEURAL NETWORK APPLICATIONS

The disclosure pertains to adaptive buffer management for supporting a dynamic tensor shape in a DNN. An apparatus for the DNN may include processor circuitry configured to:

* Determine if the tensor shape of an input tensor of an object in the DNN is dynamic and exists in a shape buffer pool.
* Run the object using a compilation result stored in the shape buffer pool when the tensor shape of the input tensor is dynamic and exists in the pool.
* Invoke the compilation procedure to perform JIT compilation for the object to obtain the compilation result when the tensor shape of the input tensor is dynamic and not in the shape buffer pool.

Key Features and Innovation:
- Adaptive buffer management for dynamic tensor shapes in a DNN.
- Utilization of a shape buffer pool to store compilation results for objects.
- JIT compilation invoked for objects with dynamic tensor shapes not in the buffer pool.

Potential Applications:
- Deep learning models with dynamically changing tensor shapes.
- Real-time processing of data with varying tensor dimensions.
- Optimization of DNN performance by reducing compilation overhead.

Problems Solved:
- Efficient management of dynamic tensor shapes in deep neural networks.
- Minimization of compilation time for objects with changing tensor shapes.
- Improved overall performance of DNNs by utilizing pre-compiled results.

Benefits:
- Faster processing of objects with dynamic tensor shapes.
- Reduction in computational overhead for compilation.
- Enhanced efficiency and performance of deep learning models.

Commercial Applications:
- AI and machine learning platforms for dynamic data processing.
- Cloud computing services for real-time analytics.
- Edge computing devices for on-the-fly model adaptation.

Questions about Adaptive Buffer Management for Dynamic Tensor Shapes:
1. How does adaptive buffer management improve the efficiency of deep neural networks?
2. What are the potential challenges in implementing adaptive buffer management for dynamic tensor shapes in DNNs?

Frequently Updated Research:
- Stay updated on advancements in adaptive buffer management techniques for dynamic tensor shapes in deep learning models.


==Original Abstract Submitted==

The disclosure relates to adaptive buffer management to support a dynamic tensor shape in a DNN. An apparatus for the DNN may include processor circuitry configured to: determine whether a tensor shape of an input tensor of an object in the DNN is dynamic and exists in a shape buffer pool; run the object by use of a compilation result for the object stored in the shape buffer pool when the tensor shape of the input tensor is dynamic and exists in the shape buffer pool; and invoke the compilation procedure to perform JIT compilation for the object so as to get the compilation result for the object when the tensor shape of the input tensor is dynamic and does not exist in the shape buffer pool.



[[Category:G06N3/08]]
[[Category:CPC_G06N3/08]]