Main Areas of Focus in the Workshop
The workshop will focus on three interconnected topics of particular relevance to Machine Learning:
- Biological Inspiration and Inductive Biases: To what extent must human-level learning be innately constrained---through bias or biology---as opposed to more fully data-driven? Recognizing which biases humans bring to learning tasks could provide insights into more sample-efficient machine learning.
- New Datasets for Comparing Humans and Machines: While machine learning algorithms and AI systems have continued to achieve human-level performance on a growing list of benchmark tasks, humans are still the gold standard in many respects. To understand the differences between human and machine strategies in more detail, large new datasets are needed to expose deeper representational and behavioral characteristics of humans.
- Robustness and Generalization: What are the underlying mechanisms that make human perception invariant/robust to distortions, rotations, scaling, and occlusion? Identifying relevant mechanisms and representations may provide insights into human perception which consequently lead to more robust AI systems.
Talks & Panels
In addition to having three sub-sessions with four speakers each (1 hour total per sub-session), we will be hosting a poster session, as well as three interactive Panel Discussions. The first panel is an Ask a Neuro / Cognitive Scientist, a Q&A session where where the audience (mostly machine learning scientists) can ask questions to expert scientists about any topic ranging from theories of human learning to representations in visual cortex and critical differences between deep neural networks and the human visual system. The second panel is targeted to international undergraduate students whose participation in machine learning venues such as NeurIPS has substantially increased in recent years. In this second panel, we will have a group of Graduate Students who are willing to share their experiences as cross-disciplinary students (e.g., neuroscience and machine learning), in addition to providing some tips and tricks on graduate school admissions. The final panel is a discussion/debate between the audience i.e., the ML community and a group of panelists/speakers (behavioral and brain scientists) who will exchange ideas on the aspects that will be crucial for developing effective artificial intelligence.