FPGA-Based AI Smart NICs for Scalable Distributed AI Training Systems

Efficient Techniques for Distributed Implementation of Search-Based AI Systems

We study the problem of exploiting parallelism from search-based AI systems on distributed machines. We propose stack-splitting, a technique for implementing orparallelism, which when coupled with appropriate scheduling strategies leads to: (i) reduced communication during distributed execution; and, (ii) distribution of larger grainsized work to processors. The modified technique can also be i...

Distributed Control for AI

This paper discusses a number of elementary problems in distributed computing and a couple of well-known algorithmic \building blocks", which are used as procedures in distributed applications. We shall not strive for completeness, as an enumeration of the many known distributed algorithms would be pointless and endless. We do not even try to touch all relevant sub-areas and problems studied in...

Improving Search-Based AI Systems

Metadatabase Meets Distributed AI

Heterogeneous Distributed Database Management Systems (HDDBMS) involve the interoperability of data sources. One approach to achieve this type of integration is to build interfaces between the diierent databases being integrated. This approach holds, for a particular case, at a speciic point in time. In this case however, the database structures need to be adapted. Such adaptation is not advisa...

AI{based Trading in Open Distributed Environments

An open distributed environment can be perceived as a service market where services are freely o ered and requested. Any infrastructure which seeks to provide appropriate mechanisms for such an environment has to include mediator functionality (i.e. a trader) that matches service requests and service o ers. Commonly, the matching process is based upon some IDL{based service type de nition, and ...