Search Results for author: Eugenio Culurciello

Found 17 papers, 9 papers with code

Capsule Network Performance with Autonomous Navigation

no code implementations8 Feb 2020 Thomas Molnar, Eugenio Culurciello

Other approaches for navigating sparse environments require intrinsic reward generators, such as the Intrinsic Curiosity Module (ICM) and Augmented Curiosity Modules (ACM).

Autonomous Navigation

Continual Reinforcement Learning in 3D Non-stationary Environments

1 code implementation24 May 2019 Vincenzo Lomonaco, Karan Desai, Eugenio Culurciello, Davide Maltoni

High-dimensional always-changing environments constitute a hard challenge for current reinforcement learning techniques.

reinforcement-learning Reinforcement Learning (RL)

Deep Predictive Coding Network with Local Recurrent Processing for Object Recognition

1 code implementation NeurIPS 2018 Kuan Han, Haiguang Wen, Yizhen Zhang, Di Fu, Eugenio Culurciello, Zhongming Liu

When unfolded over time, the recurrent processing gives rise to an increasingly deeper hierarchy of non-linear transformation, allowing a shallow network to dynamically extend itself into an arbitrarily deep network.

Image Classification Object Recognition

Deep Predictive Coding Network for Object Recognition

no code implementations ICML 2018 Haiguang Wen, Kuan Han, Junxing Shi, Yizhen Zhang, Eugenio Culurciello, Zhongming Liu

Given image input, PCN runs recursive cycles of bottom-up and top-down computation to update its internal representations and reduce the difference between bottom-up input and top-down prediction at every layer.

Image Classification Object +1

Snowflake: A Model Agnostic Accelerator for Deep Convolutional Neural Networks

no code implementations8 Aug 2017 Vinayak Gokhale, Aliasger Zaidy, Andre Xian Ming Chang, Eugenio Culurciello

Snowflake is able to achieve a computational efficiency of over 91% on modern CNN models.

Hardware Architecture

Compiling Deep Learning Models for Custom Hardware Accelerators

no code implementations1 Aug 2017 Andre Xian Ming Chang, Aliasger Zaidy, Vinayak Gokhale, Eugenio Culurciello

Given a programmable hardware accelerator with a CNN oriented custom instructions set, the compiler's task is to exploit the hardware's full potential, while abiding with the hardware constraints and maintaining generality to run different CNN models with varying workload properties.

object-detection Object Detection

CortexNet: a Generic Network Family for Robust Visual Temporal Representations

no code implementations8 Jun 2017 Alfredo Canziani, Eugenio Culurciello

In the past five years we have observed the rise of incredibly well performing feed-forward neural networks trained supervisedly for vision related tasks.

Object Recognition

An Analysis of Deep Neural Network Models for Practical Applications

4 code implementations24 May 2016 Alfredo Canziani, Adam Paszke, Eugenio Culurciello

Since the emergence of Deep Neural Networks (DNNs) as a prominent technique in the field of computer vision, the ImageNet classification challenge has played a major role in advancing the state-of-the-art.

Robust Convolutional Neural Networks under Adversarial Noise

no code implementations19 Nov 2015 Jonghoon Jin, Aysegul Dundar, Eugenio Culurciello

Recent studies have shown that Convolutional Neural Networks (CNNs) are vulnerable to a small perturbation of input called "adversarial examples".

Convolutional Clustering for Unsupervised Learning

no code implementations19 Nov 2015 Aysegul Dundar, Jonghoon Jin, Eugenio Culurciello

In this work, we propose to train a deep convolutional network based on an enhanced version of the k-means clustering algorithm, which reduces the number of correlated parameters in the form of similar filters, and thus increases test categorization accuracy.

Clustering Image Classification

Recurrent Neural Networks Hardware Implementation on FPGA

1 code implementation17 Nov 2015 Andre Xian Ming Chang, Berin Martini, Eugenio Culurciello

Recurrent Neural Networks (RNNs) have the ability to retain memory and learn data sequences.

Language Modelling

Flattened Convolutional Neural Networks for Feedforward Acceleration

2 code implementations17 Dec 2014 Jonghoon Jin, Aysegul Dundar, Eugenio Culurciello

We present flattened convolutional neural networks that are designed for fast feedforward execution.

An Analysis of the Connections Between Layers of Deep Neural Networks

1 code implementation1 Jun 2013 Eugenio Culurciello, Jonghoon Jin, Aysegul Dundar, Jordan Bates

On the other hand, in unsupervised learning, one cannot rely on back-propagation techniques to learn the connections between layers.

Cannot find the paper you are looking for? You can Submit a new open access paper.