The Principled Design of Large-Scale Recursive Neural Network Architectures--DAG-RNNs and the Protein Structure Prediction Problem

By Pierre Baldi, Gianluca Pollastri. 2003

228 citations recorded by the source corpus.

How researchers see this paper

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

Citations in the corpus

Deep learning in neural networks: An overview (2014, neutral): “DAG-RNNs (Baldi and Pollastri, 2003; Wu and Baldi, 2008) generalize BRNNs to multiple dimensions.”
NNcon: improved protein contact map prediction using 2D-recursive neural networks (2009, methodology): “2D-RNN is a 2D machine learning method designed to map 2D input information into 2D output targets (36).”
Offline Handwriting Recognition with Multidimensional Recurrent Neural Networks (2008, background): “MDRNNs, which are a special case of directed acyclic graph networks [1], generalise standard RNNs by providing recurrent connections along all spatio-temporal dimensions present in the data.”
Deep learning in bioinformatics (2016, unknown): “Research in bioinformatics using human designed features as input data such as PSSM from genomics sequences or wavelet energy from EEG signals can be understood in the same context [86, 92, 172, 176].”
Distance-Aware DAG Embedding for Proximity Search on Heterogeneous Graphs (2018, unknown): “2005) for scene labeling, (Baldi and Pollastri 2003) for protein structure prediction, and (Zhu, Sobhani, and Guo 2016) for semantic compositionality.”
Graph kernels for chemical informatics (2005, methodology): “…neural network (RNN) approach [Baldi and Chauvin, 1996, Goller and Kuchler, 1996, Sperduti and Starita, 1997, LeCun et al., 1998, Frasconi et al., 1998, Micheli et al., 2001, 2003, Baldi and Pollastri, 2003] can be viewed as a variation of the graphical model approach with Bayesian networks .”
SCRATCH: a protein structure and structural feature prediction server (2005, methodology): “CMAPpro : The contact map predictions are included as an attachment to the structural features email message.”
Offline Handwriting Recognition with Multidimensional Recurrent Neural Networks (2008, unknown): “MDRNNs, which are a special case of directed acyclic graph networks [1], generalise standard RNNs by providing recurrent connections along all spatio-temporal dimensions present in the data.”
AOGNets: Compositional Grammatical Architectures for Deep Learning (2017, background): “onnection and dense-connection adapt the sequential architecture to directed and acyclic graph (DAG) structured networks, which were explored earlier in the context of recurrent neural networks (RNN) [3,18] and ConvNets [67]. Most work focused on boosting spatial encoding and utilizing spatial dimen”
Machine Learning Methods for Protein Structure Prediction (2008, background): “[28], [70]–[72], self-organizing maps [73], and support vector”