Method and apparatus for intercollaboration of cognitive models
View Patent ↗A first cognitive instance receives information about other cognitive instances and from this compiles a cognitive community map that associates individual ones of the other cognitive instances with specific capabilities of said respective other cognitive instances. The first cognitive instance stores that map in a local memory of the first cognitive instance; and when the first cognitive instance executes a cognitive computing program it checks the cognitive community map for at least one of the specific capabilities relevant for executing that program to address/satisfy a user request that caused the program to execute. In various embodiment these cognitive instances share their respective cognitive capabilities via cognitive capability maps, which may be refreshed in their local memories for example by sending a broadcast message. Thus any given cognitive instance can identify cognitive peers with the capabilities most relevant to assist itself in solving a given problem/request.
1. A method comprising:
receiving at a first cognitive instance over an electronic communication network information about other cognitive instances;
compiling from the received information a cognitive community map that associates individual ones of the other cognitive instances with specific capabilities of said respective other cognitive instances;
storing the cognitive community map in a local memory of the first cognitive instance;
when executing a cognitive computing program at the first cognitive instance, checking the cognitive community map for at least one of the specific capabilities relevant for executing the cognitive computing program;
selecting a cognitive peer from among the other cognitive instances based on the at least one of the specific capabilities relevant for executing the cognitive computing program;
sending to the cognitive peer a query about run time capabilities of the cognitive peer; and
in response to determining that the run time capabilities of the cognitive peer are suitable, locking at least one aspect of the cognitive peer to assure the run time capabilities of the cognitive peer in service of the first cognitive instance.
2. The method according to claim 1 , wherein:
the compiling comprises refreshing a previously-stored cognitive community map; and
the receiving is in reply to a query message sent from the first cognitive instance to all of the other cognitive instances that were listed on the previously-stored cognitive community map.
3. The method according to claim 2 , wherein the information received at the first cognitive instance comprises cognitive capability maps from the other cognitive instances.
4. The method according to claim 1 , wherein the specific capabilities of at least some of the other cognitive instances includes at least:
wisdom facet information;
expert footprint information;
cost to compute information;
proximity to the first cognitive instance information; and
run-time non-functional requirements.
5. The method according to claim 4 , wherein at least one of:
the wisdom facet information includes:
level identifier,
format processing ability,
paraphrase distance,
accuracy credentials,
precision credentials and
security credentials;
or
the run-time non-functional requirements includes:
static run-time requirements,
dynamic run-time requirements,
availability and
response time.
6. A method comprising:
receiving at a first cognitive instance over an electronic communication network information about other cognitive instances;
compiling from the received information a cognitive community map that associates individual ones of the other cognitive instances with specific capabilities of said respective other cognitive instances;
storing the cognitive community map in a local memory of the first cognitive instance;
when executing a cognitive computing program at the first cognitive instance, checking the cognitive community map for at least one of the specific capabilities relevant for executing the cognitive computing program;
when executing the cognitive computing program at the first cognitive instance a query from a user of the first cognitive instance is identified as being beyond a confidence level of the cognitive computing program to address; and
checking the cognitive community map for at least one of the specific capabilities relevant for executing the cognitive computing program comprises:
using pre-established cognitive community access rules and the at least one of the specific capabilities relevant for executing the cognitive computing program to select a cognitive peer from among the other cognitive instances; and
invoking the cognitive peer to assist in addressing the query.
7. The method according to claim 1 , further comprising:
replying to a poll, received at the first cognitive instance from one of the other cognitive instances or from a cognitive capability hub entity, with at least one of:
information about capabilities of the first cognitive instance; or
the cognitive capability map.
8. An apparatus comprising:
one or more memories comprising non-transitory computer-readable code; and
one or more processors,
wherein the one or more processors are configured, in response to execution of the computer-readable code, to cause the apparatus to perform actions comprising:
receive at a first cognitive instance over an electronic communication network information about other cognitive instances;
compile from the received information a cognitive community map that associates individual ones of the other cognitive instances with specific capabilities of said respective other cognitive instances;
store the cognitive community map in a local memory of the first cognitive instance;
when executing a cognitive computing program at the first cognitive instance, check the cognitive community map for at least one of the specific capabilities relevant for executing the cognitive computing program;
reply to a poll, received at the first cognitive instance from one of the other cognitive instances or from a cognitive capability hub entity, with at least one of:
information about capabilities of the first cognitive instance; or
the cognitive capability map.
9. The apparatus according to claim 8 , wherein:
the cognitive community map is compiled by refreshing a previously-stored cognitive community map; and
the information about other cognitive instances is received in reply to a query message sent from the first cognitive instance to all of the other cognitive instances that were listed on the previously-stored cognitive community map.
10. The apparatus according to claim 9 , wherein the information received at the first cognitive instance comprises cognitive capability maps from the other cognitive instances.
11. The apparatus according to claim 8 , wherein the specific capabilities of at least some of the other cognitive instances includes at least:
wisdom facet information;
expert footprint information;
cost to compute information;
proximity to the first cognitive instance information; and
run-time non-functional requirements.
12. The apparatus according to claim 11 , wherein at least one of:
the wisdom facet information includes:
level identifier,
format processing ability,
paraphrase distance,
accuracy credentials,
precision credentials and
security credentials;
or
the run-time non-functional requirements includes:
static run-time requirements,
dynamic run-time requirements,
availability and
response time.
13. The apparatus according to claim 8 , wherein:
when executing the cognitive computing program at the first cognitive instance a query from a user of the first cognitive instance is identified as being beyond a confidence level of the cognitive computing program to address; and
the cognitive community map is checked for at least one of the specific capabilities relevant for executing the cognitive computing program by:
using pre-established cognitive community access rules and the at least one of the specific capabilities relevant for executing the cognitive computing program to select a cognitive peer from among the other cognitive instances; and
invoking the cognitive peer to assist in addressing the query.
14. The apparatus according to claim 8 , the actions further comprising:
select a cognitive peer from among the other cognitive instances based on the at least one of the specific capabilities relevant for executing the cognitive computing program;
send to the cognitive peer a query about run time capabilities of the cognitive peer; and
in response to determining that the run time capabilities of the cognitive peer are suitable, lock at least one aspect of the cognitive peer to assure the run time capabilities of the cognitive peer in service of the first cognitive instance.