In contemporary cognitive science, psychological processes are primarily investigated and characterized in an ad hoc fashion with a deliberate disregard for, or little appeal to, rational grounds, i.e., what makes them normatively justified. Algorithmic Cognitive Science (ACS) is a research program aiming at inspecting cognitive processes through the algorithmic lens, a view strongly advocated in Theoretical Computer Science (TCS). The research agenda of ACS can be articulated in simple terms as follows: To appeal to core topics in TCS (formalization, axiomatization, design and analysis of algorithms, data structures, computational complexity theory, distributed computing, etc.) in our investigation and characterization of psychological processes. Much like Algorithmic Game Theory (AGT) that sets out to study strategic environments (as a subject matter of economics) using the formal tools of TCS, ACS seeks to formally characterize the algorithmic foundations of human cognition using the analytical tools of TCS. This page is meant to provide a list of algorithmic concepts and analytical tools that, in my view, may contribute to the research agenda of ACS.*
-
-
- Anytime algorithms (see Nobandegani, 2017, § 3)
- Maximally-informative subgraph (Nobandegani, 2017, § 3)
- Scale-invariance (see Nobandegani, 2017, § 2)
- Symmetry and self-symmetry (see Nobandegani, 2017, § 2)
- Lifted inference (see Nobandegani, 2017, § 2)
- Distributed computing (see Nobandegani, 2017, § 5)
- Structural Minimality (see Nobandegani, 2017, § 4)
- Minimalist-optimal importance distribution for Importance Sampling (see Nobandegani et al., 2018)
- Parameterized complexity and fixed-parameter tractability
-
*Are you interested in Algorithmic Cognitive Science too? Let’s get in touch: ardavan.salehinobandegani@mail.mcgill.ca
Ardavan S. Nobandegani 