Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userDeep Learning Models of the Retinal Response to Natural Scenes
132
0
0.0
(
0
)
Ask ChatGPT about the research
No Arabic abstract
A central challenge in neuroscience is to understand neural computations and circuit mechanisms that underlie the encoding of ethologically relevant, natural stimuli. In multilayered neural circuits, nonlinear processes such as synaptic transmission and spiking dynamics present a significant obstacle to the creation of accurate computational models of responses to natural stimuli. Here we demonstrate that deep convolutional neural networks (CNNs) capture retinal responses to natural scenes nearly to within the variability of a cells response, and are markedly more accurate than linear-nonlinear (LN) models and Generalized Linear Models (GLMs). Moreover, we find two additional surprising properties of CNNs: they are less susceptible to overfitting than their LN counterparts when trained on small amounts of data, and generalize better when tested on stimuli drawn from a different distribution (e.g. between natural scenes and white noise). Examination of trained CNNs reveals several properties. First, a richer set of feature maps is necessary for predicting the responses to natural scenes compared to white noise. Second, temporally precise responses to slowly varying inputs originate from feedforward inhibition, similar to known retinal mechanisms. Third, the injection of latent noise sources in intermediate layers enables our model to capture the sub-Poisson spiking variability observed in retinal ganglion cells. Fourth, augmenting our CNNs with recurrent lateral connections enables them to capture contrast adaptation as an emergent property of accurately describing retinal responses to natural scenes. These methods can be readily generalized to other sensory modalities and stimulus ensembles. Overall, this work demonstrates that CNNs not only accurately capture sensory circuit responses to natural scenes, but also yield information about the circuits internal structure and function.
rate research
Read More
Natural images follow statistics inherited by the structure of our physical (visual) environment. In particular, a prominent facet of this structure is that images can be described by a relatively sparse number of features. To investigate the role of this sparseness in the efficiency of the neural code, we designed a new class of random textured stimuli with a controlled sparseness value inspired by measurements of natural images. Then, we tested the impact of this sparseness parameter on the firing pattern observed in a population of retinal ganglion cells recorded ex vivo in the retina of a rodent, the Octodon degus. These recordings showed in particular that the reliability of spike timings varies with respect to the sparseness with globally a similar trend than the distribution of sparseness statistics observed in natural images. These results suggest that the code represented in the spike pattern of ganglion cells may adapt to this aspect of the statistics of natural images.
How natural communication sounds are spatially represented across the inferior colliculus, the main center of convergence for auditory information in the midbrain, is not known. The neural representation of the acoustic stimuli results from the interplay of locally differing input and the organization of spectral and temporal neural preferences that change gradually across the nucleus. This raises the question how similar the neural representation of the communication sounds is across these gradients of neural preferences, and whether it also changes gradually. Multi-unit cluster spike trains were recorded from guinea pigs presented with a spectrotemporally rich set of eleven species-specific communication sounds. Using cross-correlation, we analyzed the response similarity of spiking activity across a broad frequency range for similarly and differently frequency-tuned neurons. Furthermore, we separated the contribution of the stimulus to the correlations to investigate whether similarity is only attributable to the stimulus, or, whether interactions exist between the multi-unit clusters that lead to correlations and whether these follow the same representation as the response similarity. We found that similarity of responses is dependent on the neurons spatial distance for similarly and differently frequency-tuned neurons, and that similarity decreases gradually with spatial distance. Significant neural correlations exist, and contribute to the response similarity. Our findings suggest that for multi-unit clusters in the mammalian inferior colliculus, the gradual response similarity with spatial distance to natural complex sounds is shaped by neural interactions and the gradual organization of neural preferences.
Thalamic relay cells fire action potentials that transmit information from retina to cortex. The amount of information that spike trains encode is usually estimated from the precision of spike timing with respect to the stimulus. Sensory input, however, is only one factor that influences neural activity. For example, intrinsic dynamics, such as oscillations of networks of neurons, also modulate firing pattern. Here, we asked if retinal oscillations might help to convey information to neurons downstream. Specifically, we made whole-cell recordings from relay cells to reveal retinal inputs (EPSPs) and thalamic outputs (spikes) and analyzed these events with information theory. Our results show that thalamic spike trains operate as two multiplexed channels. One channel, which occupies a low frequency band (<30 Hz), is encoded by average firing rate with respect to the stimulus and carries information about local changes in the image over time. The other operates in the gamma frequency band (40-80 Hz) and is encoded by spike time relative to the retinal oscillations. Because these oscillations involve extensive areas of the retina, it is likely that the second channel transmits information about global features of the visual scene. At times, the second channel conveyed even more information than the first.
Models of neural responses to stimuli with complex spatiotemporal correlation structure often assume that neurons are only selective for a small number of linear projections of a potentially high-dimensional input. Here we explore recent modeling approaches where the neural response depends on the quadratic form of the input rather than on its linear projection, that is, the neuron is sensitive to the local covariance structure of the signal preceding the spike. To infer this quadratic dependence in the presence of arbitrary (e.g. naturalistic) stimulus distribution, we review several inference methods, focussing in particular on two information-theory-based approaches (maximization of stimulus energy or of noise entropy) and a likelihood-based approach (Bayesian spike-triggered covariance, extensions of generalized linear models). We analyze the formal connection between the likelihood-based and information-based approaches to show how they lead to consistent inference. We demonstrate the practical feasibility of these procedures by using model neurons responding to a flickering variance stimulus.
We present a theoretical study aiming at model fitting for sensory neurons. Conventional neural network training approaches are not applicable to this problem due to lack of continuous data. Although the stimulus can be considered as a smooth time dependent variable, the associated response will be a set of neural spike timings (roughly the instants of successive action potential peaks) which have no amplitude information. A recurrent neural network model can be fitted to such a stimulus-response data pair by using maximum likelihood estimation method where the likelihood function is derived from Poisson statistics of neural spiking. The universal approximation feature of the recurrent dynamical neuron network models allow us to describe excitatory-inhibitory characteristics of an actual sensory neural network with any desired number of neurons. The stimulus data is generated by a Phased Cosine Fourier series having fixed amplitude and frequency but a randomly shot phase. Various values of amplitude, stimulus component size and sample size are applied in order to examine the effect of stimulus to the identification process. Results are presented in tabular form at the end of this text.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
