This tutorial covers advanced concept of energy based models. The lecture is a part of the Associative memories modules of the the Deep Learning Course at CDS, a course that covered the latest techniques in deep learning and representation learning, focusing on supervised and unsupervised deep learning, embedding methods, metric learning, convolutional and recurrent nets, with applications to computer vision, natural language understanding, and speech recognition.
This lecture provides an introduction to the problem of speech recognition using neural models, emphasizing the CTC loss for training and inference when input and output sequences are of different lengths. It also covers the concept of beam search for use during inference, and how that procedure may be modeled at training time using a Graph Transformer Network. It is a part of the Deep Learning Course at CDS, a course that covered the latest techniques in deep learning and representation learning, focusing on supervised and unsupervised deep learning, embedding methods, metric learning, convolutional and recurrent nets, with applications to computer vision, natural language understanding, and speech recognition. Prerequisites for this module include: Modules 1 - 5 of this course and Introduction to Data Science or a Graduate Level Machine Learning.
This lecture covers the concepts of the architecture and convolution of traditional convolutional neural networks, the characteristics of graph and graph convolution, and spectral graph convolutional neural networks and how to perform spectral convolution, as well as the complete spectrum of Graph Convolutional Networks (GCNs), starting with the implementation of Spectral Convolution through Spectral Networks. It then provides insights on applicability of the other convolutional definition of Template Matching to graphs, leading to Spatial networks. This lecture is a part of the Deep Learning Course at CDS, a course that covered the latest techniques in deep learning and representation learning, focusing on supervised and unsupervised deep learning, embedding methods, metric learning, convolutional and recurrent nets, with applications to computer vision, natural language understanding, and speech recognition. Prerequisites for this module include: Modules 1 - 5 of this course and Introduction to Data Science or a Graduate Level Machine Learning.
This tutuorial covers the concept of Graph convolutional networks and is a part of the Deep Learning Course at CDS, a course that covered the latest techniques in deep learning and representation learning, focusing on supervised and unsupervised deep learning, embedding methods, metric learning, convolutional and recurrent nets, with applications to computer vision, natural language understanding, and speech recognition. Prerequisites for this module include: Modules 1 - 5 of this course and Introduction to Data Science or a Graduate Level Machine Learning.
This lecture covers the concept of model predictive control and is a part of the Deep Learning Course at CDS, a course that covered the latest techniques in deep learning and representation learning, focusing on supervised and unsupervised deep learning, embedding methods, metric learning, convolutional and recurrent nets, with applications to computer vision, natural language understanding, and speech recognition. Prerequisites for this module include: Models 1-6 of this course and Introduction to Data Science or a Graduate Level Machine Learning.
This lecture covers the concepts of emulation of kinematics from observations and training a policy. It is a part of the Deep Learning Course at CDS, a course that covered the latest techniques in deep learning and representation learning, focusing on supervised and unsupervised deep learning, embedding methods, metric learning, convolutional and recurrent nets, with applications to computer vision, natural language understanding, and speech recognition. Prerequisites for this module include: Models 1-6 of this course and Introduction to Data Science or a Graduate Level Machine Learning.
This lecture covers the concept of predictive policy learning under uncertainty and is a part of the Deep Learning Course at CDS, a course that covered the latest techniques in deep learning and representation learning, focusing on supervised and unsupervised deep learning, embedding methods, metric learning, convolutional and recurrent nets, with applications to computer vision, natural language understanding, and speech recognition. Prerequisites for this module include: Models 1-6 of this course and Introduction to Data Science or a Graduate Level Machine Learning.
This lecture covers the concepts of gradient descent, stochastic gradient descent, and momentum. It is a part of the Deep Learning Course at CDS, a course that covered the latest techniques in deep learning and representation learning, focusing on supervised and unsupervised deep learning, embedding methods, metric learning, convolutional and recurrent nets, with applications to computer vision, natural language understanding, and speech recognition. Prerequisites for this module include: Models 1-7 of this course and Introduction to Data Science or a Graduate Level Machine Learning.
This lecture covers the concepts of gradient descent, stochastic gradient descent, and momentum. It is a part of the Deep Learning Course at CDS, a course that covered the latest techniques in deep learning and representation learning, focusing on supervised and unsupervised deep learning, embedding methods, metric learning, convolutional and recurrent nets, with applications to computer vision, natural language understanding, and speech recognition. Prerequisites for this module include: Models 1-7 of this course and Introduction to Data Science or a Graduate Level Machine Learning.
Peer Herholz gives a tour of how popular virtualization tools like Docker and Singularity are playing a crucial role in improving reproducibility and enabling high-performance computing in neuroscience.
Blake Richards gives an introduction to deep learning, with a perspective via inductive biases and emphasis on correctly matching deep learning to the right research questions.
The lesson was presented in the context of the BrainHack School 2020.
Félix-Antoine Fortin from Calcul Québec gives an introduction to high-performance computing with the Compute Canada network, first providing an overview of use cases for HPC and then a hand-on tutorial. Though some examples might seem specific to the Calcul Québec, all computing clusters in the Compute Canada network share the same software modules and environments.
The lesson was given in the context of the BrainHack School 2020.