Content provided by Pranav Sharma, the author of the paper LPar – A Distributed Multi Agent Platform for Building Polyglot, Omni Channel and Industrial Grade Natural Language Interfaces.
The goal of serving and delighting customers in a personal and near human like manner is very high on automation agendas of most Enterprises. Last few years, have seen huge progress in Natural Language Processing domain which has led to deployments of conversational agents in many enterprises. Most of the current industrial deployments tend to use Monolithic Single Agent designs that model the entire knowledge and skill of the Domain. While this approach is one of the fastest to market, the monolithic design makes it very hard to scale beyond a point. There are also challenges in seamlessly leveraging many tools offered by sub fields of Natural Language Processing and Information Retrieval in a single solution. The sub fields that can be leveraged to provide relevant information are, Question and Answer system, Abstractive Summarization, Semantic Search, Knowledge Graph etc. Current deployments also tend to be very dependent on the underlying Conversational AI platform (open source or commercial), which is a challenge as this is a fast evolving space and no one platform can be considered future proof even in medium term of 3-4 years. Lately, there is also work done to build multi agent solutions that tend to leverage a concept of master agent. While this has shown promise, this approach still makes the master agent in itself difficult to scale. To address these challenges, we introduce LPar, a distributed multi agent platform for large scale industrial deployment of polyglot, diverse and inter-operable agents. The asynchronous design of LPar supports dynamically expandable domain. We also introduce multiple strategies available in the LPar system to elect the most suitable agent to service a customer query.
What’s New: A federated and scalable multi agent design that enables agents built using different commercial and open source Conversational AI tools to co-exist and service the user in a completely transparent manner. The system leverages pod based domain driven design and serving store construct to provide the capability to dynamically expand the domain of the overall system by adding new agents at the run time without effecting the overall system. The design also enables usage of tools providing different NLP capabilities like Abstractive Summarization, Question and Answer System, Semantic Search etc along side Conversation AI agents. This provides the Agent Developers and product teams the flexibility they need to use the most suitable tool to model their specific use case. For example, by leveraging the proposed system, they can use a self hosted Conversational AI tool for modelling data sensitive intents like KYC but can leverage managed cloud Conversational AI tools for intents like simple FAQs where Data Privacy is not a concern to get desired productivity benefits. The system provides a context sharing and session management services that can be leveraged by Agent developers to share and retrieve context across agents. The system also provides a unique broadcast and select algorithm to choose the most suitable agent to service any particular customer query.
Key Insights: In an enterprise context, leveraging a federated multi agent approach is more suitable compared to using monolithic design approaches that model all the intents in a single agent. It is also important to avoid master agent based approaches in a multi agent system as the master agents themselves become a giant monolith over a period of time and become a bottleneck to dynamically scale the domain of the system. The design proposed in this paper enables multiple agents built using diverse tools to effectively co-exist and share context. This provides product and engineering teams flexibility to choose NLP tools and platforms based on the use cases. The federated nature of the system supports capability to dynamically expand domain of the system by adding agents at the run time and also increases maintainability with ease of debugging issues due to domain driven design of agents.
The paper LPar – A Distributed Multi Agent Platform for Building Polyglot, Omni Channel and Industrial Grade Natural Language Interfaces is on arXiv.
Meet the authors Pranav Sharma from Cognizant Worldwide Limited.
Share Your Research With Synced Review
Share My Research is Synced’s new column that welcomes scholars to share their own research breakthroughs with over 1.5M global AI enthusiasts. Beyond technological advances, Share My Research also calls for interesting stories behind the research and exciting research ideas. Share your research with us by clicking here.
0 comments on “LPar – A Distributed Multi Agent Platform for Building Polyglot, Omni Channel and Industrial Grade Natural Language Interfaces”