Current research directions of the Brain Engineering Lab include but not limited to:

• Multi-scale Brain Simulation.
• Neural Computation and Neural Network Modeling.
• Micro-scale Brain Structure Reconstruction.
• Brain-inspired Computing.
• Brain Knowledge Bases.
• Cognitive and Neural Robotics.

Cognitive Brain Modeling Group
The Cognitive Brain Modeling Group, founded in June, 2013 (previously as the Neural Computation Group), focuses on modeling and simulating the cognitive brain and developing brain-inspired intelligent systems.

In order to simulate the cognitive brain at multiple scales, efforts in the form of data and literatures for understanding how the brain works need to be synthesized and well organized in a structured way. The efforts in this direction are under the framework of the Linked Neuron Data (LND) project. This project is an effort for extracting and linking Neuroscience data and knowledge from multiple scales and multiple data sources together. The LND platform provides services for neuroscience knowledge extraction, structured neuron data representation, neuron data integration, visualization, analysis, semantic search and reasoning over the integrated Neuron data. More than 5 million knowledge triples about the brain were extracted and synthesized from various sources (including unstructured source such as all the scientific papers from PubMed, and structured source such as NIF and NeuroLex). The Brain Association Graph, which contains the relationships among cognitive functions, brain diseases and brain building blocks at multiple scales (including more than 1 million knowledge triples extracted from PubMed articles and Wikipedia pages, etc.) is the core of LND that connects various understanding of the brain together.

The core effort of the Cognitive Brain Modeling Group is to simulate the cognitive brain at multiple scales to explore and understand how the brain works. CASIA Brain Simulation System simulates the cognitive brain from ions, neurons, neural micro-circuits and meso-circuits, all the way to brain regions, macro circuits and cognitive behaviors. The morphologies and the activities of the brain building blocks at multiple scales are acquired with different methodologies and equipments (such as ion channel and patch-clamp techniques, two-photon calcium imaging, fMRI, EEG, etc.) and synthesized from various sources at multiple scales, and simulation is strictly based on the observed data and principles. The long term goal of CASIA Brain Simulation effort is to decode the mechanisms and principles of human intelligence and develop Brain-inspired intelligent systems. Recently, the group has just established a cat V2 orientation selection simulation system and a mouse hippocampus simulation system. From the function point of view, the CASIA Brain simulator has already been able to perform object recognition, inductive reasoning, working memory, reinforcement learning and knowledge processing tasks. These tasks are executed based on the CASIA Brain simulator, which coordinates different simulated brain regions to achieve various cognitive abilities.

Brain-inspired Information Processing Group

The Brain-inspired Information Processing Group investigate on the next generation information processing theories, techniques, and applications inspired by the information processing principles from the brain. Its current goal is to build a Multi-modal Cognitive Machine with vision, audition, language processing, thinking capabilities. It has great potentials in the field of large-scale, multi-modal data and information processing.

Neuro-robotics Group

The core task of Neuro-Robotics Group (NRG) is to do research and development on the next generation of intelligent robot system, by imitating and learning the mechanics of the human neural system. Currently, NRG is conducting the CASIA Motor Brain Project and studying the mechanics of conscious movement and subconscious movement simultaneously. Up to now, the basic model has been established and a hardware platform is being built to test and exhibit the results. A series of related papers have been published on IEEE Transactions, and NRG is invited by Frontiers in Computational Neuroscience to organize a special issue on this field.