Ralf Wessel

Motion of a point particle, rotational motion, oscillation, gravitation and central forces, Lagrangian and Hamiltonian formulation.

Concepts and techniques of physics are applied to study the functioning of neurons and neuronal circuits in the brain. Neurons and neural systems are modeled at two levels: (i) at the physical level, in terms of the electrical and chemical signals that are generated and transmitted and (ii) at the information-processing level, in terms of the computational tasks performed. Specific topics include: neuronal electrophysiology, neural codes, neural plasticity, sensory processing, neural network architectures and learning algorithms, and neural networks as dynamical and statistical systems.

How do the eyes capture an image and convert it to neural messages that ultimately result in visual experience? This lecture and demonstration course will cover the physics of how we see. The course is addressed to physics, premedical, and life-sciences students with an interest in biophysics. Topics include physical properties of light, evolution of the eyes, image formation in the eye, image sampling with an array of photoreceptors, transducing light into electrical signals, color coding, retinal organization, computing with nerve cells, compressing the 3-D world into optic nerve signals, inferring the 3-D world from optic nerve signals, biomechanics of eye movement, engineered vision in machines. The functional impact of biophysical mechanisms for visual experience will be illustrated with psychophysical demonstrations.

This laboratory course consists of "table-top" experiments in biological physics that are designed to introduce the student to concepts, methods, and biological model systems in biophysics. Most experiments combine experimentation with computer simulations. The list of available experiments includes electrophysiology, human bioelectricity, optical tweezers, ultrasonic imaging, mass spectrometer, and viscosity measurements.

What can Deep Learning teach us about the Brain? What can Artificial Intelligence in Neural Networks teach us about the functioning of Natural Intelligence in Brains? What can Computational Neuroscience teach us about the functioning of Artificial Intelligence in Neural Networks? Do the mechanisms of Artificial Intelligence in deep networks serve as a framework for the investigation of Natural Intelligence in evolved brains? Deep Neural Networks (DNNs) have revolutionized Machine Learning and Artificial Intelligence. These networks have recently found their way back into computational neuroscience. DNNs reach human-level performance in certain tasks. Importantly, DNNs can display certain characteristics of brain function that cannot be captured with shallow handcrafted models. With this, DNNs offer an intriguing novel framework that may enable computational neuroscientists to address fundamental questions concerning complexity and neural computation in brains. In this course we will evaluate the tantalizing hypothesis that DNNs can serve as a tool to understand aspects of brain function, thus moving closer towards an understanding of both, natural and artificial intelligence. The course will be organized around 15 key papers illustrating major ideas of DNNs and computational neuroscience.