Search Results - "Liao, Qianli"

Complexity control by gradient descent in deep networks by Poggio, Tomaso, Liao, Qianli, Banburski, Andrzej

“…Overparametrized deep networks predict well, despite the lack of an explicit complexity control during training, such as an explicit regularization term. For…”
Get full text

View-Tolerant Face Recognition and Hebbian Learning Imply Mirror-Symmetric Neural Tuning to Head Orientation by Leibo, Joel Z., Liao, Qianli, Anselmi, Fabio, Freiwald, Winrich A., Poggio, Tomaso

“…The primate brain contains a hierarchy of visual areas, dubbed the ventral stream, which rapidly computes object representations that are both specific for…”
Get full text

Dynamics in Deep Classifiers Trained with the Square Loss: Normalization, Low Rank, Neural Collapse, and Generalization Bounds by Xu, Mengjia, Rangamani, Akshay, Liao, Qianli, Galanti, Tomer, Poggio, Tomaso

“…We overview several properties-old and new-of training overparameterized deep networks under the square loss. We first consider a model of the dynamics of…”
Get full text

The Invariance Hypothesis Implies Domain-Specific Regions in Visual Cortex by Leibo, Joel Z, Liao, Qianli, Anselmi, Fabio, Poggio, Tomaso

“…Is visual cortex made up of general-purpose information processing machinery, or does it consist of a collection of specialized modules? If prior knowledge,…”
Get full text

Theoretical issues in deep networks by Poggio, Tomaso, Banburski, Andrzej, Liao, Qianli

“…While deep learning is successful in a number of applications, it is not yet well understood theoretically. A theoretical characterization of deep learning…”
Get full text

Why and When Can Deep-but Not Shallow-networks Avoid the Curse of Dimensionality： A Review by Poggio, Tomaso, Mhaskar, Hrushikesh, Rosasco, Lorenzo, Miranda, Brando, Liao, Qianli

“…The paper reviews and extends an emerging body of theoretical results on deep learning including the conditions under which it can be exponentially better than…”
Get full text

Hierarchically Local Tasks and Deep Convolutional Networks by Deza, Arturo, Banburski, Andrzej, Liao, Qianli, Poggio, Tomaso

Get full text

Compression of Deep Neural Networks for Image Instance Retrieval by Chandrasekhar, Vijay, Jie Lin, Qianli Liao, Morere, Olivier, Veillard, Antoine, Lingyu Duan, Poggio, Tomaso

“…Image instance retrieval is the problem of retrieving images from a database which contain the same object. Convolutional Neural Network (CNN) based…”
Get full text

The Invariance Hypothesis Implies Domain-Specific Regions in Visual Cortex: e1004390 by Leibo, Joel Z, Liao, Qianli, Anselmi, Fabio, Poggio, Tomaso

“…Is visual cortex made up of general-purpose information processing machinery, or does it consist of a collection of specialized modules? If prior knowledge,…”
Get full text

Explicit regularization and implicit bias in deep network classifiers trained with the square loss by Poggio, Tomaso, Liao, Qianli

“…Deep ReLU networks trained with the square loss have been observed to perform well in classification tasks. We provide here a theoretical justification based…”
Get full text

Theory II: Landscape of the Empirical Risk in Deep Learning by Liao, Qianli, Poggio, Tomaso

“…Previous theoretical work on deep learning and neural network optimization tend to focus on avoiding saddle points and local minima. However, the practical…”
Get full text

Theoretical Issues in Deep Networks: Approximation, Optimization and Generalization by Poggio, Tomaso, Banburski, Andrzej, Liao, Qianli

“…While deep learning is successful in a number of applications, it is not yet well understood theoretically. A satisfactory theoretical characterization of deep…”
Get full text

Bridging the Gaps Between Residual Learning, Recurrent Neural Networks and Visual Cortex by Liao, Qianli, Poggio, Tomaso

“…We discuss relations between Residual Networks (ResNet), Recurrent Neural Networks (RNNs) and the primate visual cortex. We begin with the observation that a…”
Get full text

Hierarchically Compositional Tasks and Deep Convolutional Networks by Deza, Arturo, Liao, Qianli, Banburski, Andrzej, Poggio, Tomaso

“…The main success stories of deep learning, starting with ImageNet, depend on deep convolutional networks, which on certain tasks perform significantly better…”
Get full text

Biologically-plausible learning algorithms can scale to large datasets by Xiao, Will, Chen, Honglin, Liao, Qianli, Poggio, Tomaso

“…The backpropagation (BP) algorithm is often thought to be biologically implausible in the brain. One of the main reasons is that BP requires symmetric weight…”
Get full text

A Surprising Linear Relationship Predicts Test Performance in Deep Networks by Liao, Qianli, Miranda, Brando, Banburski, Andrzej, Hidary, Jack, Poggio, Tomaso

“…Given two networks with the same training loss on a dataset, when would they have drastically different test losses and errors? Better understanding of this…”
Get full text

Streaming Normalization: Towards Simpler and More Biologically-plausible Normalizations for Online and Recurrent Learning by Liao, Qianli, Kawaguchi, Kenji, Poggio, Tomaso

“…We systematically explored a spectrum of normalization algorithms related to Batch Normalization (BN) and propose a generalized formulation that simultaneously…”
Get full text

Learning Functions: When Is Deep Better Than Shallow by Mhaskar, Hrushikesh, Liao, Qianli, Poggio, Tomaso

“…While the universal approximation property holds both for hierarchical and shallow networks, we prove that deep (hierarchical) networks can approximate the…”
Get full text

How Important is Weight Symmetry in Backpropagation? by Liao, Qianli, Leibo, Joel Z, Poggio, Tomaso

“…Gradient backpropagation (BP) requires symmetric feedforward and feedback connections -- the same weights must be used for forward and backward passes. This…”
Get full text

Theory III: Dynamics and Generalization in Deep Networks by Banburski, Andrzej, Liao, Qianli, Miranda, Brando, Rosasco, Lorenzo, De La Torre, Fernanda, Hidary, Jack, Poggio, Tomaso

“…The key to generalization is controlling the complexity of the network. However, there is no obvious control of complexity -- such as an explicit…”
Get full text