Theano: Deep Learning on GPUs with Python

By James Bergstra, Frédéric Bastien, Olivier Breuleux, Pascal Lamblin, Razvan Pascanu, Olivier Delalleau, Guillaume Desjardins, David Warde-Farley, Ian J. Goodfellow, Arnaud Bergeron and 1 more. 2012

280 citations recorded by the source corpus.

Subfields: Large Language Models

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Large Language Models

Citations in the corpus

Theano: new features and speed improvements (2012, methodology): “Bergstra et al. (2010, 2011), as well as Theano’s website 1 have more in-depth descriptions and examples.”
NICE: Non-linear Independent Components Estimation (2014, methodology): “n Courville, Kyle Kastner, Dustin Webb, Li Yao and Aaron Van den Oord for discussions and feedback. Vincent Dumoulin provided code for visualization. We are grateful towards the developers of Theano (Bergstra et al., 2011; Bastien et al., 2012) and Pylearn2 (Goodfellow et al., 2013), and for the com”
Deep Learning of Representations: Looking Forward (2013, methodology): “Note also that sparse coding can be generalized into the spike-and-slab sparse coding algorithm (Goodfellow et al., 2012), in which MAP inference is replaced by variational inference, and that was used to win the NIPS 2011 transfer learning challenge (Goodfellow et al., 2011).”
Deep Neural Networks Based Recognition of Plant Diseases by Leaf Image Classification (2016, background): “ere are several well-known state-of-the-art deep learning frameworks, such as Python library eano [37] andmachine learning library that extends Lua, Torch7 [38].”
CRADLE: Cross-Backend Validation to Detect and Localize Bugs in Deep Learning Libraries (2019, methodology): “as TensorFlow (Google) [9], Theano [10], and CNTK (Microsoft) [11], implement the same algorithms, e.”
On-chip training of recurrent neural networks with limited numerical precision (2017, methodology): “The program is implemented with Theano library in python [18].”
Deep learning in robotics: a review of recent research (2017, background): “now available: Tensorflow has recently surged in popularity [100]. Theano is a Python-based platform that provides General Purpose Graphical Processing Unit (GPGPU)- parallelization for deep learning [101]. Several popular toolkits build on top of Theano, including as Lasagne and Pylearn2 [102]. Ker”
Deep learning for healthcare applications based on physiological signals: A review (2018, unknown): “Graphics Processing Units (GPUs) can be used to reduce the network latency [162].”
End-to-end Deep Learning of Optimization Heuristics (2017, methodology): “We implemented the model using Keras, with TensorFlow [16] and Theano [17] backends.”
SPINDLE: SPINtronic Deep Learning Engine for large-scale neuromorphic computing (2014, methodology): “Several research eﬀorts have been devoted to realizing eﬃcient implementations of DLNs using multi-core processors, graphics processing units, and hardware accelerators [6–8].”
ActiVis: Visual Exploration of Industry-Scale Deep Neural Network Models (2017, background): “Deep neural network models are often represented as computation graphs (DAGs) (as in many deep learning frameworks like Caffe [16], TensorFlow [1], Theano [6]).”
Comparative Study of Deep Learning Software Frameworks (2015, methodology): “Developers have constantly improved these frameworks by adding more features (e.g. by adding support for diﬀerent types of convolution algorithms) and speed improvements to attract more users and foster research [4, 7, 21, 1, 13].”