Deep Neural Networks Reveal a Gradient in the Complexity of Neural Representations across the Ventral Stream

Güçlü, Umut; Gerven, Marcel A. J. van

doi:10.1523/jneurosci.5023-14.2015

Cited by 860 publications

(558 citation statements)

References 53 publications

Supporting

Mentioning

520

Contrasting

Order By: Relevance

“…Of course, we might also expect some differences in the development of object recognition across species, especially between chickens and humans. For instance, humans have much larger visual systems than chickens, which may allow humans to achieve greater levels of abstraction across the successive levels of the visual cortex [36,37,44]. Further, chickens, unlike humans, are mobile from birth and immediately able to explore their environment.…”

Section: Discussionmentioning

confidence: 99%

The development of newborn object recognition in fast and slow visual worlds

Wood

2016

Proc. R. Soc. B.

View full text Add to dashboard Cite

Object recognition is central to perception and cognition. Yet relatively little is known about the environmental factors that cause invariant object recognition to emerge in the newborn brain. Is this ability a hardwired property of vision? Or does the development of invariant object recognition require experience with a particular kind of visual environment? Here, we used a high-throughput controlled-rearing method to examine whether newborn chicks (Gallus gallus) require visual experience with slowly changing objects to develop invariant object recognition abilities. When newborn chicks were raised with a slowly rotating virtual object, the chicks built invariant object representations that generalized across novel viewpoints and rotation speeds. In contrast, when newborn chicks were raised with a virtual object that rotated more quickly, the chicks built viewpoint-specific object representations that failed to generalize to novel viewpoints and rotation speeds. Moreover, there was a direct relationship between the speed of the object and the amount of invariance in the chick's object representation. Thus, visual experience with slowly changing objects plays a critical role in the development of invariant object recognition. These results indicate that invariant object recognition is not a hardwired property of vision, but is learned rapidly when newborns encounter a slowly changing visual world.

show abstract

Section: Discussionmentioning

confidence: 99%

The development of newborn object recognition in fast and slow visual worlds

Wood

2016

Proc. R. Soc. B.

View full text Add to dashboard Cite

show abstract

“…A number of BOLD-MRI studies have revealed that the neural activation's in early areas of visual cortex show the best correspondence with the early layers of DNNs and that higher-tier cortical areas show the best correspondence with higher-tier DNN layers (Eickenberg, Gramfort, Varoquaux, & Thirion, 2017;Gü çlü & van Gerven, 2015). MEG/EEG studies have furthermore shown that early layers of DNNs have a peak explained variance that is earlier than higher-tier DNN layers (Cichy, Khosla, Pantazis, Torralba, & Oliva, 2016;Ramakrishnan, Scholte, Smeulders, & Ghebreab, 2016).…”

Section: 1mentioning

confidence: 99%

Visual pathways from the perspective of cost functions and multi-task deep neural networks

Scholte

Losch

Ramakrishnan

et al. 2018

Cortex

View full text Add to dashboard Cite

Deep learning Cost functions RepresentationsVisual processing a b s t r a c t Vision research has been shaped by the seminal insight that we can understand the higher-tier visual cortex from the perspective of multiple functional pathways with different goals. In this paper, we try to give a computational account of the functional organization of this system by reasoning from the perspective of multi-task deep neural networks. Machine learning has shown that tasks become easier to solve when they are decomposed into subtasks with their own cost function. We hypothesize that the visual system optimizes multiple cost functions of unrelated tasks and this causes the emergence of a ventral pathway dedicated to vision for perception, and a dorsal pathway dedicated to vision for action.To evaluate the functional organization in multi-task deep neural networks, we propose a method that measures the contribution of a unit towards each task, applying it to two networks that have been trained on either two related or two unrelated tasks, using an identical stimulus set. Results show that the network trained on the unrelated tasks shows a decreasing degree of feature representation sharing towards higher-tier layers while the network trained on related tasks uniformly shows high degree of sharing.We conjecture that the method we propose can be used to analyze the anatomical and functional organization of the visual system and beyond. We predict that the degree to which tasks are related is a good descriptor of the degree to which they share downstream cortical-units.

show abstract

“…The structure and operation of a neuron-the basic building block of ANN-have been described on several occasions (Guclu et al 2015). Briefly, it is a cascaded design, from an input layer to an output layer, where functional blocks are sandwiched either between input and hidden, or hidden and output layers, where each layer may comprise a specific number of neurons.…”

Section: Artificial Neural Network Modelmentioning

confidence: 99%

Modeling electrical properties for various geometries of antidots on a superconducting film

et al. 2017

View full text Add to dashboard Cite

Electrical properties, specifically critical current density, of a superconducting film carry a substantial importance in superconductivity. In this work, we measure and study the current-voltage curves for a superconducting Nb film with various geometries of antidots to tune the critical current. We carry out the measurements on a commercially available physical property measurement system to obtain these so-called transport measurements. We show that each of the used geometries exhibits a vastly different critical current, due to which repeatedly performing the measurements independently for each geometry becomes indispensable. To circumvent this monotonous measurement procedure, we also propose a framework based on artificial neural networks to predict the curves for different geometries using a small subset of measurements, and facilitate extrapolation of these curves over a wide range of parameters including temperature and magnetic field. The predicted curves are then crosschecked using the physical measurements; our results suggest a negligible mean-squared error-in the order of 10 À9 .

show abstract

Deep Neural Networks Reveal a Gradient in the Complexity of Neural Representations across the Ventral Stream

Cited by 860 publications

References 53 publications

The development of newborn object recognition in fast and slow visual worlds

The development of newborn object recognition in fast and slow visual worlds

Visual pathways from the perspective of cost functions and multi-task deep neural networks

Modeling electrical properties for various geometries of antidots on a superconducting film

Contact Info

Product

Resources

About