Network composition module for a bayesian neuromorphic compiler
Described is a system for specifying control of a device based on a Bayesian network model. The system includes a Bayesian neuromorphic compiler having a network composition module having probabilistic computation units (PCUs) arranged in a hierarchical composition containing multi-level dependencies. The Bayesian neuromorphic compiler receives a Bayesian network model as input and produces a spiking neural network topology and configuration that implements the Bayesian network model. The network composition module learns conditional probabilities of the Bayesian network model. The system computes a conditional probability and controls a device based on the computed conditional probability.
1. A system for specifying control of a device based on a Bayesian network model, the system comprising:
a Bayesian neuromorphic compiler, having a network composition module comprising a plurality of probabilistic computation units (PCUs) arranged in a hierarchical composition containing multi-level dependencies, embodied on a non-transitory computer-readable medium, the Bayesian neuromorphic compiler comprising instructions to cause one or more processors to perform operations of:
receiving a Bayesian network model as input;
producing a spiking neural network topology and configuration that implements the Bayesian network model; and
using the network composition module, learning conditional probabilities of the Bayesian network model;
computing at least one conditional probability; and
controlling a device based on the at least one computed conditional probability.
2. The system as set forth in claim 1 , wherein controlling the device includes causing the device to perform a maneuvering operation based on the at least one computed conditional probability.
3. The system as set forth in claim 2 , wherein the device is an autonomous mobile platform, and the maneuvering operation is a collision avoidance operation.
4. The system as set forth in claim 1 , wherein the PCUs are neuromorphic computing circuits configured to compute conditional probabilities.
5. The system as set forth in claim 1 , wherein the network composition module arranges the plurality of PCUs such that it learns the conditional probabilities of the Bayesian network model and, subsequently, infers at least one conditional probability pertaining to a query for the Bayesian network model.
6. A computer implemented method for specifying control of a device based on a Bayesian network model, the method comprising acts of:
using a Bayesian neuromorphic compiler having a network composition module comprising a plurality of probabilistic computation units (PCUs) arranged in a hierarchical composition containing multi-level dependencies, embodied on a non-transitory computer-readable medium, the Bayesian neuromorphic compiler comprising instructions to cause one or more processors to perform operations of:
receiving a Bayesian network as input;
producing a spiking neural network topology and configuration that implements the Bayesian network model;
using the network composition module, learning conditional probabilities of the Bayesian network model;
computing at least one conditional probability; and
controlling a device based on the at least one computed conditional probability.
7. The method as set forth in claim 6 , wherein controlling the device includes causing the device to perform a maneuvering operation based on the at least one computed conditional probability.
8. The method as set forth in claim 7 , wherein the device is an autonomous mobile platform, and the maneuvering operation is a collision avoidance operation.
9. The method as set forth in claim 6 , wherein the PCUs are neuromorphic computing circuits configured to compute conditional probabilities.
10. The method as set forth in claim 6 , wherein the network composition module arranges the plurality of PCUs such that it learns the conditional probabilities of the Bayesian network model and, subsequently, infers at least one conditional probability pertaining to a query for the Bayesian network model.
11. A computer program product for specifying control of a device based on a Bayesian network model, the computer program product comprising:
a Bayesian neuromorphic compiler, having a network composition module comprising a plurality of probabilistic computation units (PCUs) arranged in a hierarchical composition containing multi-level dependencies, embodied on a non-transitory computer-readable medium, the Bayesian neuromorphic compiler comprising instructions to cause one or more processors to perform operations of:
receiving a Bayesian network model as input;
producing a spiking neural network topology and configuration that implements the Bayesian network model; and
using the network composition module, learning conditional probabilities of the Bayesian network model;
computing at least one conditional probability; and
controlling a device based on the at least one computed conditional probability.
12. The computer program product as set forth in claim 11 , wherein controlling the device includes causing the device to perform a maneuvering operation based on the at least one computed conditional probability.
13. The computer program product as set forth in claim 12 , wherein the device is an autonomous mobile platform, and the maneuvering operation is a collision avoidance operation.
14. The computer program product as set forth in claim 11 , wherein the PCUs are neuromorphic computing circuits configured to compute conditional probabilities.
15. The computer program product as set forth in claim 11 , wherein the network composition module arranges the plurality of PCUs such that it learns the conditional probabilities of the Bayesian network model and, subsequently, infers at least one conditional probability pertaining to a query for the Bayesian network model.