A5078357982- Neural dynamics and brain function
- Visual perception and processing mechanisms
- Neuroscience and Music Perception
- Cell Image Analysis Techniques
- Retinal Development and Disorders
- Blind Source Separation Techniques
- Speech and Audio Processing
- Advanced Adaptive Filtering Techniques
- Hearing Loss and Rehabilitation
University of Oxford
2017-2025
Neurons in sensory cortex are tuned to diverse features natural scenes. But what determines which neurons become selective to? Here we explore the idea that neuronal selectivity is optimized represent recent past best predict immediate future inputs. We tested this hypothesis using simple feedforward neural networks, were trained next few moments of video or audio clips The networks developed receptive fields closely matched those real cortical different mammalian species, including oriented...
Visual neurons respond selectively to features that become increasingly complex from the eyes cortex. Retinal prefer flashing spots of light, primary visual cortical (V1) moving bars, and those in higher areas favor like textures. Previously, we showed V1 simple cell tuning can be accounted for by a basic model implementing temporal prediction – representing predict future sensory input past (Singer et al., 2018). Here, show hierarchical application capture how properties change across at...
We investigate how the neural processing in auditory cortex is shaped by statistics of natural sounds. Hypothesising that (A1) represents structural primitives out which sounds are composed, we employ a statistical model to extract such components. The input cochleagrams approximate non-linear transformations sound undergoes from outer ear, through cochlea nerve. Cochleagram components do not superimpose linearly, but rather according rule can be approximated using max function. This...
Visual neurons respond selectively to specific features that become increasingly complex in their form and dynamics from the eyes cortex. Retinal prefer localized flashing spots of light, primary visual cortical (V1) moving bars, those higher areas, such as middle temporal (MT) cortex, favor like textures. Whether there are general computational principles behind this diversity response properties remains unclear. To date, no single normative model has been able account for hierarchy tuning...
Neurons in primary visual cortex (V1) show a remarkable functional specificity their pre- and postsynaptic partners. Recent work has revealed variety of wiring biases describing how the short- long-range connections V1 neurons relate to tuning properties. However, it is less clear whether these connectivity rules are based on some underlying principle cortical organization. Here, we that emerges naturally recurrent neural network optimized predict upcoming sensory inputs for natural stimuli....
Neurons in sensory cortex are tuned to diverse features natural scenes. But what determines which neurons become selective to? Here we explore the idea that neuronal selectivity is optimised represent recent past of input best predict immediate future inputs. We tested this hypothesis using simple feedforward neural networks, were trained next few video or audio frames clips The networks developed receptive fields closely matched those real cortical neurons, including oriented spatial tuning...
Almost half a billion people world-wide suffer from disabling hearing loss. While aids can partially compensate for this, large proportion of users struggle to understand speech in situations with background noise. Here, we present deep learning-based algorithm that selectively suppresses noise while maintaining signals. The restores intelligibility aid the level control subjects normal hearing. It consists network is trained on custom database noisy signals and further optimized by neural...