Recent advancements in artificial intelligence have serious people discussing or even claiming the existence of artificial general intelligence (AGI). In particular, large language models (LLMs) demonstrate a major milestone. Are they a major step towards AGI or a statistical trick? During 'Machine Intelligence', we explore these topics from numerous perspectives.
Surveys! You've taken them. You can't help but critique them. This season explores topics in survey design, methodology, interpretation, and audience sourcing.
The internet is ubiquitous. It's mostly paid for by advertising. Our season explores the algorithms, technology, platforms, and perspectives on the world of real time digital advertising.
The pandemic was a change point in many ways. Remote work and remote learning become more commonplace. Today organizations are re-evaluating the pros and cons of remote and in-person work. This Data Skeptic mini-series explores the data available about challenges being physically distributed.
The unsupervised learning algorithm k-means requires one input parameter: the k. Given k and your data, the algorithm returns an answer which labels each element of your data to a cluster. The algorithm finds k 'centroids' and points are assigned to the nearest centroid. The point of the algorithm is to find the best set of centroids given the data.
When data is collected sequentially or a measurement is taken periodically in time, the result is time series data. Time series is typically separated into the topics of analysis and forecasting. This season of Data Skeptic explores the tools, methodologies, and use cases for time series problems.
Consensus means many different things in different contexts. It has relevance in distributed systems, database design, and voting algorithms. This season is an exploration of all these areas.
The unreasonable impact of machine learning techniques demonstrate that they are here to stay. That being the case, it is critically that when an important decision is made by an algorithm, it be possible for the people affected by that decision to understand how the algorithm arrived at it's conclusion. This season is a broad exploration of explainability and interpretability techniques for AI and ML.
While fake news may not be a recent invention, it is one that has certainly one that has taken on a new context given the rise of social media and the effectiveness of efforts for motivated parties to mislead others. This season is a collection of interviews with researchers looking at this topic.
Amongst the hardest problems in computer science are challenges around getting machines to understand and generate natural language. This season is an exploration of current research, applications, and thought leaders with something to say about artifical intelligence.
Alan Turing's empirical test of machine intelligence known as the 'Imitation Game' still stands as the benchmark for artifical general intelligence. While recent advancements in deep learning techniques seem to make this goal more attainable, there are still many questions about just how far we are from this goal.