Abstract
Draculab is a neural simulator with a particular use scenario: firing rate units with delayed connections, using custom-made unit and synapse models, possibly controlling simulated physical systems. Draculab also has a particular design philosophy. It aims to blur the line between users and developers. Three factors help to achieve this: a simple design using Python's data structures, extensive use of standard libraries, and profusely commented source code. This paper is an introduction to Draculab's architecture and philosophy. After presenting some example networks it explains basic algorithms and data structures that constitute the essence of this approach. The relation with other simulators is discussed, as well as the reasons why connection delays and interaction with simulated physical systems are emphasized.
Highlights
When faced with a new project, modelers in computational neuroscience must decide whether to program their simulations from scratch, or to use one of the existing neural simulators
Research in firing rate neural networks has brought a large number of results
Many of the early results produced the Parallel Distributed Processing Paradigm (Rumelhart and Mcclelland, 1986), eventually leading to the more recent “deep learning” trend (Schmidhuber, 2015). Because of their feedforward connectivity and lack of continuous dynamics these models have deviated from biological plausibility, and it is unclear whether their computation style is anything like biological brains
Summary
When faced with a new project, modelers in computational neuroscience must decide whether to program their simulations from scratch, or to use one of the existing neural simulators. Simulations where neural controllers interact with a physical system, implementing closed-loop control Besides satisfying these requirements, Draculab aims to have a simple interface, but still provide total control over the simulation to experienced Python users. For experienced Python users, Draculab aims to give ample control over the simulation, so it feels like they wrote the simulator themselves (but without going through the hassle of writing and testing it). This comes from a straightforward implementation, profusely commented source code, and standard libraries such as Scipy
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call