System and method for removing biases within a distributable model
A system for improving a distributable model with distributed data is provided, comprising a network-connected distributable model source configured to serve instances of a distributable model; and a directed computation graph module configured to receive at least an instance of the distributable model from the network-connected computing system, create a cleansed dataset from data stored in the memory with at least biases within the data stored in memory corrected, train the instance of the distributable model with the cleansed dataset, and generate an update report based at least in part by updates to the instance of the distributable model.
1. A system for improving a distributable model with distributed data, comprising:
a network-connected distributable model source comprising a memory, a processor, and a plurality of programming instructions stored in the memory thereof and operable on the processor thereof, wherein the programmable instructions, when operating on the processor, cause the processor to:
serve instances of a distributable model; and
a directed computational graph module comprising a memory, a processor, and a plurality of programming instructions stored in the memory thereof and operable on the processor thereof, wherein the programmable instructions, when operating on the processor, cause the processor to:
receive at least an instance of the distributable model from the network-connected computing system;
operate a directed computational graph comprising a plurality of data transformation nodes;
remove a plurality of data biases within the instance of the distributable model using at least a portion of the plurality of data transformation nodes;
create a cleansed dataset based on the instance of the distributable model and the removed biases;
wherein the biases comprise trends exhibited within the data;
wherein the biases are automatically intelligently weighted and corrected by the distributed computational graph to generalize the dataset for use in a distributable model;
train the instance of the distributable model with the cleansed dataset; and
generate an update report based at least in part by updates to the instance of the distributable model.
2. The system of claim 1 , wherein at least a portion of the cleansed dataset is data that has had sensitive information removed.
3. The system of claim 1 , wherein the at least a portion of the update report is used by the network-connected distributable model source to improve the distributable model.
4. The system of claim 1 , wherein at least a portion of the data stored in memory is medical-related data.
5. The system of claim 1 , wherein at least a portion of the data stored in memory is crime-related data.
6. The system of claim 1 , wherein at least a portion of the data stored in memory is banking-related data.
7. The system of claim 3 , wherein the network-connected distributable model source classifies an incoming update report based at least in part by geographical origin of the incoming update report.
8. The system of claim 7 , wherein the update report is used to improve a distributable model specific to the geographical origin.
9. The system of claim 8 , wherein the distributable model specific to the geographical origin is not improvable with update reports from restricted geographical origins.
10. A method for improving a distributable model with distributed data, comprising the steps of:
(a) serving instances of a distributable model with a network-connected distributable model source;
(b) receiving at least an instance of the distributable model from the network-connected computing system with a directed computational graph module;
(c) operating a directed computational graph comprising a plurality of data transformation nodes;
(d) removing a plurality of data biases within the instance of the distributable model using at least a portion of the plurality of data transformation nodes;
(e) creating a cleansed dataset based on the instance of the distributable model and the removed biases;
wherein the biases comprise trends exhibited within the data;
wherein the biases are automatically intelligently weighted and corrected by the distributed computational graph to generalize the dataset for use in a distributable model;
(f) training the instance of the distributable model with the cleansed dataset with the directed computation graph module; and
(g) generating an update report based at least in part by updates to the instance of the distributable model with the directed computation graph module.
11. The method of claim 10 , wherein at least a portion of the cleansed dataset is data that has had sensitive information removed.
12. The method of claim 10 , wherein the at least a portion of the update report is used by the network-connected distributable model source to improve the distributable model.
13. The method of claim 10 , wherein at least a portion of the data stored in memory is medical-related data.
14. The method of claim 10 , wherein at least a portion of the data stored in memory is crime-related data.
15. The method of claim 10 , wherein at least a portion of the data stored in memory is banking-related data.
16. The method of claim 12 , wherein the network-connected distributable model source categorizes an incoming update report based at least in part by geographical origin of the incoming update report.
17. The method of claim 16 , wherein the update report is used to improve a distributable model specific to the geographical origin.
18. The method of claim 17 , wherein the distributable model specific to the geographical origin is not improvable with update reports from other geographical origins.