Adaptive quantization for neural networks

1. A processor configured for adaptive quantization in an artificial neural network (ANN), the processor comprising:

processor circuitry configured to calculate a distribution of ANN information;

processor circuitry configured to select a quantization function for each layer of the ANN from a set of quantization functions based on the distribution, wherein a first layer of the ANN has a different selected quantization function than a second layer of the ANN;

processor circuitry configured to apply the quantization function to the ANN information to generate quantized ANN information;

processor circuitry configured to load the quantized ANN information into the ANN; and

processor circuitry configured to generate an output based on the quantized ANN information.

2. The processor of claim 1 , further comprising processor circuitry configured to, on a condition that the output does not meet an acceptability criterion:

recalculate the distribution of ANN information; and

reselect the quantization function from the set of quantization functions based on the recalculated distribution.

3. The processor of claim 1 , wherein the ANN information comprises a set of training data.

4. The processor of claim 1 , wherein the ANN information comprises a plurality of link weights.

5. The processor of claim 4 , further comprising processor circuitry configured to:

calculate a distribution of link weights for each of a plurality of layers of the ANN;

select a quantization function to the plurality of link weights for each of the plurality of layers of the ANN based on each distribution; and

apply the respective quantization function to the link weights for each of the plurality of layers.

6. The processor of claim 4 , further comprising processor circuitry configured to:

calculate a distribution of link weights for each of a plurality of subsets of layers of the ANN;

select a quantization function to the plurality of link weights for each of the plurality of subsets of layers of the ANN based on each distribution; and

apply the respective quantization function to the link weights for each of the plurality of subsets of layers.

7. The processor of claim 1 , further comprising processor circuitry configured to apply a heuristic to the output and a known correct output, and on a condition that the heuristic is satisfied, to:

recalculate the distribution of ANN information; and

reselect the quantization function from the set of quantization functions based on the recalculated distribution.

8. The processor of claim 1 , wherein the different selected quantization function is selected for a group of layers which includes the first layer.

9. A method for adaptive quantization in an artificial neural network (ANN), comprising:

calculating a distribution of ANN information;

selecting a quantization function for each layer of the ANN from a set of quantization functions based on the distribution, wherein a first layer of the ANN has a different selected quantization function than a second layer of the ANN;

applying the quantization function to the ANN information to generate quantized ANN information;

loading the quantized ANN information into the ANN; and

generating an output based on the quantized ANN information.

10. The method of claim 9 , further comprising, on a condition that the output does not meet an acceptability criterion:

recalculating the distribution of ANN information; and

reselecting the quantization function from the set of quantization functions based on the recalculated distribution.

11. The method of claim 9 , wherein the ANN information comprises a set of training data.

12. The method of claim 9 , wherein the ANN information comprises a plurality of link weights.

13. The method of claim 12 , further comprising:

calculating a distribution of link weights for each of a plurality of layers of the ANN;

selecting a quantization function to the plurality of link weights for each of the plurality of layers of the ANN based on each distribution; and

applying the respective quantization function to the link weights for each of the plurality of layers.

14. The method of claim 12 , further comprising:

calculating a distribution of link weights for each of a plurality of subsets of layers of the ANN;

selecting a quantization function to the plurality of link weights for each of the plurality of subsets of layers of the ANN based on each distribution; and

applying the respective quantization function to the link weights for each of the plurality of subsets of layers.

15. The method of claim 9 , further comprising applying a heuristic to the output and a known correct output on a condition that the output does not meet an acceptability criterion, and on a condition that the heuristic is satisfied:

recalculate the distribution of ANN information; and

reselect the quantization function from the set of quantization functions based on the recalculated distribution.

16. The method of claim 9 , wherein the different selected quantization function is selected for a group of layers which includes the first layer.

17. A non-transitory computer-readable medium comprising instructions thereon which when executed by a processor configured for adaptive quantization in an artificial neural network (ANN), cause circuitry of the processor to:

calculate a distribution of ANN information;

select a quantization function for each layer of the ANN from a set of quantization functions based on the distribution, wherein a first layer of the ANN has a different selected quantization function than a second layer of the ANN;

apply the quantization function to the ANN information to generate quantized ANN information;

load the quantized ANN information into the ANN; and

generate an output based on the quantized ANN information.

18. The non-transitory computer-readable medium of claim 17 , further comprising instructions thereon which when executed by a processor configured for adaptive quantization in an artificial neural network (ANN), cause circuitry of the processor to, on a condition that the output does not meet an acceptability criterion:

recalculate the distribution of ANN information; and

reselect the quantization function from the set of quantization functions based on the recalculated distribution.

19. The non-transitory computer-readable medium of claim 17 , wherein the ANN information comprises a set of training data.

20. The non-transitory computer-readable medium of claim 17 , wherein the ANN information comprises a plurality of link weights.

21. The non-transitory computer-readable medium of claim 20 , further comprising instructions thereon which when executed by a processor configured for adaptive quantization in an artificial neural network (ANN), cause circuitry of the processor to:

calculate a distribution of link weights for each of a plurality of layers of the ANN;

select a quantization function to the plurality of link weights for each of the plurality of layers of the ANN based on each distribution; and

apply the respective quantization function to the link weights for each of the plurality of layers.

22. The non-transitory computer-readable medium of claim 20 , further comprising instructions thereon which when executed by a processor configured for adaptive quantization in an artificial neural network (ANN), cause circuitry of the processor to:

calculate a distribution of link weights for each of a plurality of subsets of layers of the ANN;

select a quantization function to the plurality of link weights for each of the plurality of subsets of layers of the ANN based on each distribution; and

apply the respective quantization function to the link weights for each of the plurality of subsets of layers.

23. The non-transitory computer-readable medium of claim 17 , wherein the different selected quantization function is selected for a group of layers which includes the first layer.