Organized by the non-profit research organization ContinualAI, the conference aims at speeding-up the long desired inclusive and sustainable progress of our community with an open-access, multi-timezone, 24 hours long event which brings together ideas at the intersection of machine learning, computational neuroscience, robotics and more!
The natural world is a dynamic environment that is constantly changing, requiring humans and animals to adapt. However, much research in Artificial Intelligence (AI) is focused on the static train-test paradigm with heavily curated datasets. CLAI Unconf has a broad focus beyond these static assumptions. In particular, its themes include (but are not limited to):
Navigating complex data collection systems
Understanding and describing continuous streams of data
Lifelong learning processes and generalization of knowledge beyond a specific target
Discovering new concepts in changing environments and handling partially observable information from potentially disparate data sources
Related topics from an interdisciplinary perspective
A more detailed account is provided in the Call For Papers. We look forward to your submissions at the un-conference!
Easy accessibility: virtual, multi-timezone support
Roundtables: active roundtables, not just panels of the same experts! Share your topic in advance so we can discuss it with a voting system like Reddit's.
Hands-on sessions: get your hands dirty, work collaboratively! Express your creativity: research is not only about formal research papers!
We are welcoming original contributions in the form of pre-registration articles. Pre-registration ensures scientific excellence through a two-stage submission procedure that separates the quality of scientific hypotheses from potentially negative downstream empirical analysis.
Researchers first disseminate well-articulated ideas with a thoroughly outlined experimental protocol, incorporate feedback through active dialogue at the conference, and their rigorously evaluated follow-up findings later get published in CLAI Unconf's proceedings of machine learning research (PMLR).
Thus, ideas can be presented and discussed within the community independently of conventional publication mechanisms, before ideas are potentially shut down prematurely or full-fledged publications finished necessarily. The process further guarantees that valuable insights from well formulated hypotheses with negative downstream outcomes are shared with the community.