System and method for approximating query results using neural networks
A system and method for generating approximations of query results. The method includes sending a received query to a neural network, wherein the received query is executable on a target data set; receiving from the neural network a predicted result to the received query; providing the predicted result as a first output to a device having initiated the received query; determining a real result of the query from a data set stored in the database when the predicted result is insufficiently accurate; and providing the real result as a second output to a device having initiated the received query.
1. A method for generating approximations of query results, comprising:
sending a received query from an approximation system to a neural network, wherein the received query is executable on a target data set, wherein the received query is sent from a user node and received at the approximation system, wherein the user node is a device having initiated the received query;
receiving, at the approximation system from the neural network, a predicted result to the received query;
providing the predicted result as a first output to the user node;
determining a real result of the query from a data set stored in the database when the predicted result is insufficiently accurate; and
providing the real result as a second output to the user node.
2. The method of claim 1 , further comprising:
determining if the predicted result is within a predetermined threshold of the real result.
3. The method of claim 2 , further comprising:
providing the second output when the predicted result falls outside the predetermined threshold of the real result, wherein the second output is provided to the user node after the first output is provided to the user node.
4. The method of claim 3 , further comprising:
sending the real result to the neural network when the predicted result falls outside of the predetermined threshold of the real result, wherein the real result is utilized to train the neural network.
5. The method of claim 4 , wherein training the neural network further comprises: adjusting a weight value of at least one neuron of a plurality of neurons based on a comparison between the predicated result and the real result.
6. The method of claim 1 , wherein the neural network comprises a plurality of neurons configured to generate the predicted result, and wherein each neuron is assigned a weight value.
7. The method of claim 1 , wherein the predicted result is provided without querying the database with the received query.
8. The method of claim 1 , further comprising:
querying the database with the received query to provide the real result.
9. A non-transitory computer readable medium having stored thereon instructions for causing a processing circuitry to perform a process of generating approximations of query results, the process comprising:
sending a received query from an approximation system to a neural network, wherein the received query is executable on a target data set, wherein the received query is sent from a user node and received at an approximation system, wherein the user node is a device having initiated the received query;
receiving, at the approximation system from the neural network, a predicted result to the received query;
providing the predicted result as a first output to the user node;
determining a real result of the query from a data set stored in the database when the predicted result is insufficiently accurate; and
providing the real result as a second output to the user node.
10. A system for generating approximations of query results, comprising:
a processing circuitry; and
a memory, the memory containing instructions that, when executed by the processing circuitry, configure the system to:
send a received query to a neural network, wherein the received query is executable on a target data set, wherein the received query is sent from a user node and received at the system, wherein the user node is a device having initiated the received query;
receive from the neural network a predicted result to the received query;
provide the predicted result as a first output to the user node;
determine a real result of the query from a data set stored in the database when the predicted result is insufficiently accurate; and
provide the real result as a second output to the user node.
11. The system of claim 10 , wherein the system is further configured to:
determine if the predicted result is within a predetermined threshold of the real result.
12. The system of claim 11 , wherein the system is further configured to:
provide the second output when the predicted result falls outside the predetermined threshold of the real result, wherein the second output is provided to the user node after the first output is provided to the user node.
13. The system of claim 12 , wherein the system is further configured to:
send the real result to the neural network when the predicted result falls outside of the predetermined threshold of the real result, wherein the real result is utilized to train the neural network.
14. The system of claim 13 , wherein the system is further configured to:
adjust a weight value of at least one neuron of a plurality of neurons based on a comparison between the predicated result and the real result.
15. The system of claim 10 , wherein the neural network comprises a plurality of neurons configured to generate the predicted result, and wherein each neuron is assigned a weight value.
16. The system of claim 10 , wherein the predicted result is provided without querying the database with the received query.
17. The system of claim 10 , wherein the system is further configured to:
query the database with the received query to provide the real result.
18. The system of claim 10 , wherein the system further comprises a neural computer.
19. The system of claim 18 , wherein the neural computer comprises at least one of: a von-Neumann multiprocessor, a graphical processing unit (GPU), a vector processor, an array processor, and a tensor processing unit.
20. A system for generating approximations of query results, comprising:
an approximation server;
a neural network; and
wherein the approximation server is configured to:
send a received query to a neural network, wherein the received query is executable on a target data set, wherein the received query is sent from a user node and received at the approximation server, wherein the user node is a device having initiated the received query;
receive from the neural network a predicted result to the received query;
provide the predicted result as a first output to the user node;
determine a real result of the query from the data set, when the predicted result is insufficiently accurate; and
provide the real result as a second output to the user node.