Using distributed state machines for human-to-computer dialogs with automated assistants to protect private data
Techniques are described related to for generating/distributing state machines that are implemented within a security zone to obtain private information from one or more resources within the security zone. In various implementations, an automated assistant client implemented by processor(s) within the security zone may receive a free form natural language query (“FFNLQ”) that is answerable using private information available from resource(s) within the security zone. Data indicative of the FFNLQ may be provided to a semantic processor outside of the security zone, and the online semantic processor may return a state machine that is implemented by processor(s) within the security zone to obtain the private information from resource(s) within the security zone. Based on the state machine and the obtained private information, natural language output may be generated and presented to convey information responsive to the FFNLQ.
1. A method implemented by one or more processors within a security zone, comprising:
receiving, at an automated assistant client implemented by one or more of the processors of a client device within the security zone, a free form natural language query from a user, wherein the free form natural language query is answerable based on private information that is available from one or more resources within the security zone, wherein the one or more resources are not stored on the client device, and wherein the security zone comprises a plurality of networked computing devices behind a firewall;
providing, via the firewall, first data indicative of the free form natural language query to a semantic processor, wherein the semantic processor is outside of the security zone and cannot directly access the one or more resources within the security zone;
receiving, via the firewall and at one or more of the processors within the security zone, from the semantic processor, second data indicative of a state machine, wherein the state machine is implementable to obtain the private information from one or more of the resources within the security zone;
implementing, by one or more of the processors within the security zone, the state machine to obtain the private information from one or more of the resources within the security zone;
generating, based on the state machine and the obtained private information, natural language output that conveys information responsive to the free form natural language query; and
providing, by the automated assistant client, at one or more output components, the natural language output.
2. The method of claim 1 , wherein the state machine comprises instructions that are performable by the automated assistant client.
3. The method of claim 1 , wherein the state machine comprises one or more slots that are to be filled with values that are obtainable from one or more of the resources within the security zone.
4. The method of claim 1 , wherein implementing the state machine comprises implementing an interpreter to execute bytecode forming the state machine.
5. The method of claim 1 , wherein the one or more resources within the security zone include one or more databases containing medical information about a population of people associated with healthcare provider that manages the security zone.
6. A system comprising one or more processors within a security zone and memory operably coupled with the one or more processors, wherein the memory stores instructions that, in response to execution of the instructions by one or more processors, cause the one or more processors to perform the following operations:
receiving, at an automated assistant client implemented by one or more of the processors of a client device within the security zone, wherein the security zone comprises a plurality of computing devices behind a firewall, a free form natural language query from a user, wherein the free form natural language query is answerable based on private information that is available from one or more resources within the security zone, wherein the one or more resources are not stored on the client device;
providing, via the firewall, first data indicative of the free form natural language query to a semantic processor, wherein the semantic processor is outside of the security zone and cannot directly access the one or more resources within the security zone;
receiving, via the firewall and at one or more of the processors within the security zone, from the semantic processor, second data indicative of a state machine, wherein the state machine is implementable to obtain the private information from one or more of the resources within the security zone;
implementing, by one or more of the processors within the security zone, the state machine to obtain the private information from one or more of the resources within the security zone;
generating, based on the state machine and the obtained private information, natural language output that conveys information responsive to the free form natural language query; and
providing, by the automated assistant client, at one or more output components, the natural language output.
7. The system of claim 6 , wherein the state machine comprises instructions that are performable by the automated assistant client.
8. The system of claim 6 , wherein the state machine comprises one or more slots that are to be filled with values that are obtainable from one or more of the resources within the security zone.
9. The system of claim 6 , wherein implementing the state machine comprises implementing an interpreter to execute bytecode forming the state machine.
10. The system of claim 6 , wherein the one or more resources within the security zone include one or more databases containing medical information about a population of people associated with healthcare provider that manages the security zone.
11. At least one non-transitory computer-readable medium comprising instructions that, in response to execution of the instructions by one or more processors within a security zone, cause the one or more processors to perform the following operations:
receiving, at an automated assistant client implemented by one or more of the processors of a client device within the security zone, a free form natural language query from a user, wherein the free form natural language query is answerable based on private information that is available from one or more resources within the security zone, wherein the one or more resources are not stored on the client device, and wherein the security zone comprises a plurality of computing devices behind a firewall;
providing, via the firewall, first data indicative of the free form natural language query to a semantic processor, wherein the semantic processor is outside of the security zone and cannot directly access the one or more resources within the security zone;
receiving, via the firewall and at one or more of the processors within the security zone, from the semantic processor, second data indicative of a state machine, wherein the state machine is implementable to obtain the private information from one or more of the resources within the security zone;
implementing, by one or more of the processors within the security zone, the state machine to obtain the private information from one or more of the resources within the security zone;
generating, based on the state machine and the obtained private information, natural language output that conveys information responsive to the free form natural language query; and
providing, by the automated assistant client, at one or more output components, the natural language output.
12. The non-transitory computer-readable medium of claim 11 , wherein the state machine comprises instructions that are performable by the automated assistant client.
13. The non-transitory computer-readable medium of claim 11 , wherein the state machine comprises one or more slots that are to be filled with values that are obtainable from one or more of the resources within the security zone.
14. The non-transitory computer-readable medium of claim 11 , wherein implementing the state machine comprises implementing an interpreter to execute bytecode forming the state machine.