Posted on

Introduction to Agent-Based Modeling

Among researchers, there is a growing interest in conceptualizing complex problems. It requires using a system framework and using systems modeling tools to explore how components of a complex problem interact. In particular, system simulation approaches are useful tools for understanding the processes and structures involved in complex problems. Also, identifying high-leverage points in the system and evaluating hypothetical interventions becomes easier.

One tool that has extensive usage in among researchers is agent-based modeling (ABM). We define traits and initial behavior rules of an agent that organize their actions and interactions. Stochasticity plays an important part in determining which agents interact and how agents make decisions.

Posted on

What Is a Transformer? — Inside Machine Learning

An Introduction to Transformers and Sequence-to-Sequence Learning for Machine Learning

New deep learning models are introduced at an increasing rate and sometimes it's hard to keep track of all the novelties. That said, one particular neural network model has proven to be especially effective for common natural language processing tasks. The model is called a Transformer and it makes use of several methods and mechanisms that I'll introduce here. The papers I refer to in the post offer a more detailed and quantitative description.

Part 1: Sequence-to-Sequence Learning and Attention

The paper 'Attention Is All You Need' describes transformers and what is called a sequence-to-sequence architecture. Sequence-to-Sequence (or Seq2Seq) is a neural net that transforms a given sequence of elements, such as the sequence of words in a sentence, into another sequence. (Well, this might not surprise you considering the name.)

Posted on

NumHub: A Wikipedia for Data

In this interview, I’m talking with Andrey Pyankov, founder of NumHub, which is a community-driven database of numbers, statistics, market research, industry metrics, and financial data. You can find numbers ranging from Google quarterly revenue to M&M's color distribution.

All the information on NumHub is gathered by a community of analysts and researchers, and you can request data be gathered for your research or presentations.

Posted on

The Largest Developer Community: A Critical View

When developers evaluate new technologies, one of the elements they often look at is the size and strength of the community surrounding that technology. “Can I get help and support from peers when needed?” It’s one of the reasons why open source technologies tend to be so popular. Conversely, technology vendors regularly signal their virtue with community numbers: “Our product is used by millions of developers, choose us!”

However, there is a reason to be critical of this line of thinking. The activity of a core group, or indeed the vendor itself, may matter more to get great support than the sheer number of users. Most technologies are not subject to network effects: they don’t become inherently more valuable when more developers adopt them. Even in open source projects, there is often only a small number of core contributors. Furthermore, vendors may bloat the numbers they report: deliberately, or simply because they don’t have good data available.