US Patent Application 17738053. OPTIMIZING COGBOT RETRAINING simplified abstract

From WikiPatents
Jump to navigation Jump to search

OPTIMIZING COGBOT RETRAINING

Organization Name

INTERNATIONAL BUSINESS MACHINES CORPORATION


Inventor(s)

Rajesh Kumar Saxena of Maharashtra (IN)

Harish Bharti of Pune (IN)

Rakesh Shinde of Pune (IN)

Sandeep Sukhija of Rajasthan (IN)

OPTIMIZING COGBOT RETRAINING - A simplified explanation of the abstract

This abstract first appeared for US patent application 17738053 titled 'OPTIMIZING COGBOT RETRAINING

Simplified Explanation

- The patent application is about the continual retraining of Cognitive Bots (CogBots) to adapt and evolve in changing environments. - The approach involves a CogBot response model that searches for potential responses and shifts its behavior accordingly. - The process starts with receiving a prompt at a CogBot retraining framework. - The prompt is analyzed to determine potential responses. - A dialogue benchmark is generated for each potential response. - A decision shift score is generated for the prompt. - The CogBot retraining framework is updated based on the decision shift score.


Original Abstract Submitted

The continual retraining of Cognitive Bots (“CogBots”) is of CogBots allows for the adaptation and evolution to ever changing environments. In this retraining process a CogBot response model continually searches the response space for potential responses in which it may shift. An approach for optimizing such retraining of CogBots may be presented herein. The approach may include receiving a prompt at a CogBot retraining framework. The approach may include analyzing the prompt and determining potential responses to the prompt. The approach may include generating a dialogue benchmark for each of the potential responses. The approach may further include generating a decision shift score for the prompt. Further, the approach may additionally include updating the CogBot retraining framework based on the generated decision shift score.