Nov 13th, 2019 - Cooperative Machine Learning Networks with Ahmed Masud of saf.ai

by Chris Conlan

The next event in our speaker series!

Bethesda Data Science Meetup Find us on Meetup.com.

Location 7200 Wisconsin Ave, Bethesda, MD 20814

Date & Time Wednesday November 13th, 2019 from 6:30 PM - 8:00 PM

Ahmed Masud, founder of cybersecurity AI firm saf.ai, will discuss the intersection of AI and decentralized networks.

His focus is on how blockchains can be leveraged to create complex co-operative machine-learning networks such as Ensemble Learning; Hierarchical Mixture of Experts; Swarm Intelligence, etc.

When we deal with structured data and deterministic transaction rules, the programming of transaction-processors is straightforward and manually verifiable. For complex transactions, potentially involving unstructured, noisy and/or incomplete data, or where dynamic decisions amongst multiples parties are required, it’s better to utilize co-operative machine-learning techniques.

The challenges with co-operative machine learning is that of coordination between learning agents. In other words, how can it be ensured that individual decisions of machine-learning agents that make up the cooperation lead to optimal decision for a group? This requires multiple agents to coordinate on their results and ensure that the best result is recorded for subsequent inferences. Ahmed explores how blockchains can provide this coordination and consensus capability to help complex cooperative machine learning networks to work towards global goals. The talk also compares and contrasts whether the use of blockchains is better than usual techniques of building such cooperative machine learning networks.

Ahmed also provides an overview on how to build an example of such a network.

Agenda:

  • 6:30 Introductions and networking
  • 7:00 Presentation
  • 7:30 Discussion
  • 8:00 Wrap-up

Blog

Data science discussions and general updates from Conlan Scientific.

Events

Your guide to events we host and sponsor around the D.C. area.

Team

We believe good data scientists are great software developers. Our team of professionals is ready to solve your financial data challenges.