IDEAS SoCal Data Science Conference 2017

Human beings have been doing research on Artificial Intelligence for decades, and the algorithms that are commonly used today to implement AI products are not new. It is with the large volume of data available today, along with the computing power, we are finally able to implement the methodology to achieve more accurate results, and build AI products. This talk will go over the basics on machine learning and deep learning for the purpose of AI, and how Big Data can improve the performance.

As Deep Learning architectures grow in popularity, companies are beginning to use it with their existing data to gain deeper insights. BigDL enables companies with existing Spark clusters to run Deep Learning jobs where the data sits. You no longer have to export your data outside of Spark. Real-time deep learning pipelines are now possible too! In this talk, Dave will explore the common Neural Network models used by companies who already have Hadoop or Spark clusters.

Political campaigns in the U.S. have made extensive use of persuasion modeling in recent election cycles. These same modeling approaches are now making their way into commercial applications. In this presentation, I discuss political persuasion modeling, and present case studies from recent political campaigns. I'll then turn to potential commercial applications, and discuss best practices for data security, modeling, validation, and evaluation.

Data without analytics are wasted resources. Analytics without a modern data architecture is useless. With next-gen capabilities like machine learning and automation emerging, how do you set yourself up for success?

Many enterprises are finding success with an architecture that has a data lake at the core because a data lake offers the agility and scalability that is required for big data. However, not all data lakes are able to deliver analytics at the speed needed to make a disruptive difference among their competition.

With a majority of Hadoop implementations failing to make it to production, it is critical to add a big data management platform to operationalize, automate and optimize your data lake for success.

AI and ML opportunities and challenges in Pharmaceutical and Life Sciences, focus on use cases in drug discovery, clinical development, and real-world evidence.

The increasing need for clustering in several scientific domains has inevitably driven the creation of innovative algorithms, each designed to perform more efficiently in certain applications. More specifically, in many applications, the data entities involved can be portrayed effectively by a graph as a collection of nodes and edges. One of the most established algorithms for graph clustering problems is the Markov Cluster Algorithm (MCL).

When dealing with large and complex datasets, the underlying graphs can easily reach proportions that independent computing systems are inadequate to deal with. Additionally, the graphs encountered are typically sparse: the number of edges is far smaller than might be possible in a fully-connected graph. Consequently, there is a concrete need for algorithms that are designed to handle sparse graph clustering utilizing distributed computing resources.

Our motivation was the development of a distributed architecture, able to accommodate large and sparse graphs, to actualize the MCL and R-MCL algorithm. The Apache Spark framework was chosen due to its ability to utilize distributed resources and its proven track record.

Although Spark is a framework capable of handling massive datasets, it currently does not provide rich support for computation with sparse matrices and sparse graphs. Hence, methods have been implemented to enable the exploitation of sparse adjacency matrices in distributed sparse matrix multiplication, a critical component of MCL. The proposed solution can handle arbitrarily large inputs, provide almost linear speed-up with the addition of computational resources and output results directly comparable to the non-distributed reference MCL implementation.

As a data scientist at the City of Los Angeles, we see a novel form of big data- that is, a novel variety of data.This talk will analyze three projects done at the city.
1) Predicting Displacement
2) Easing Traffic
3) Targeting Audits

Space is the final frontier whose exploration requires disrupting technology advancement in many areas including AI. In this talk we will present a historical flashback of AI's application in the past and current space exploration, explain its unique technical challenges, and describe some viable use cases and emerging opportunities of future space exploration enabled by AI as a game-changing technology.