AIL721: Difference between revisions
| [checked revision] | [checked revision] |
Prashantt492 (talk | contribs) Creating course page via bot |
Bot: wrap bare course codes in wikilinks |
||
| Line 4: | Line 4: | ||
| credits = 4 | | credits = 4 | ||
| credit_structure = 3-0-2 | | credit_structure = 3-0-2 | ||
| pre_requisites = Any one of ELL409 / ELL784 / AIL701 / COL774 | | pre_requisites = Any one of [[ELL409]] / [[ELL784]] / [[AIL701]] / [[COL774]] | ||
| overlaps = COL775 approx. 80%. APL745 approx. 40%. | | overlaps = [[COL775]] approx. 80%. [[APL745]] approx. 40%. | ||
}} | }} | ||
== AIL721 : Deep Learning == | == AIL721 : Deep Learning == | ||
/ COL 341 [COL341, COL774, ELL409, ELL784] < 15% Introduction to representation learning, limitations of classical ML methods Introduction to DNN Universal approximation theorem, mathematical foundations and introduction to deep architectures, Activation functions, Learning DNN, gradient descent, Computational Graphs and Back Propagation, Regularization - Bias-Variance trade-off, Norm-penalty Early stopping Regularization: Bagging, Boosting, Ada-boost, Dropout, Batch Norm and other regularization techniques Optimization algorithms – SGD, Momentum, RMSProp, Adaptive gradient algorithms (ADAgrad, ADAM etc.) Convolutional Neural Networks (CNNs), Architectures few applications such as Object Localization Detection and Segmentation, Action Classification, sequential Models, BPTI, Problems with Recurrent neural Networks (RNNs), Long Short Term Memory (LSTM) / Gated Recurrent Units (GRUs), Bidirectional LSTMs (BLTSMs), Applications of RNNs to NLP, Language Model Unsupervised representational learning Auto Encoders and its variants, Variational inference and Deep learning: Variational Auto Encoders (VAE) and its variants, Generative Adversarial Networks Graph Convolutional networks (GCNs), Graph Attention Networks (GATs), Deep Reinforcement Learning, Advanced Topics: Fairness and Explainability in DL Adversarial Learning Neuro-Symbolic Reasoning, Neural Architecture Search. | / COL 341 [[[COL341]], [[COL774]], [[ELL409]], [[ELL784]]] < 15% Introduction to representation learning, limitations of classical ML methods Introduction to DNN Universal approximation theorem, mathematical foundations and introduction to deep architectures, Activation functions, Learning DNN, gradient descent, Computational Graphs and Back Propagation, Regularization - Bias-Variance trade-off, Norm-penalty Early stopping Regularization: Bagging, Boosting, Ada-boost, Dropout, Batch Norm and other regularization techniques Optimization algorithms – SGD, Momentum, RMSProp, Adaptive gradient algorithms (ADAgrad, ADAM etc.) Convolutional Neural Networks (CNNs), Architectures few applications such as Object Localization Detection and Segmentation, Action Classification, sequential Models, BPTI, Problems with Recurrent neural Networks (RNNs), Long Short Term Memory (LSTM) / Gated Recurrent Units (GRUs), Bidirectional LSTMs (BLTSMs), Applications of RNNs to NLP, Language Model Unsupervised representational learning Auto Encoders and its variants, Variational inference and Deep learning: Variational Auto Encoders (VAE) and its variants, Generative Adversarial Networks Graph Convolutional networks (GCNs), Graph Attention Networks (GATs), Deep Reinforcement Learning, Advanced Topics: Fairness and Explainability in DL Adversarial Learning Neuro-Symbolic Reasoning, Neural Architecture Search. | ||
Latest revision as of 16:21, 14 April 2026
| AIL721 | |
|---|---|
| Deep Learning | |
| Credits | 4 |
| Structure | 3-0-2 |
| Pre-requisites | Any one of ELL409 / ELL784 / AIL701 / COL774 |
| Overlaps | COL775 approx. 80%. APL745 approx. 40%. |
AIL721 : Deep Learning
/ COL 341 [[[COL341]], COL774, ELL409, ELL784] < 15% Introduction to representation learning, limitations of classical ML methods Introduction to DNN Universal approximation theorem, mathematical foundations and introduction to deep architectures, Activation functions, Learning DNN, gradient descent, Computational Graphs and Back Propagation, Regularization - Bias-Variance trade-off, Norm-penalty Early stopping Regularization: Bagging, Boosting, Ada-boost, Dropout, Batch Norm and other regularization techniques Optimization algorithms – SGD, Momentum, RMSProp, Adaptive gradient algorithms (ADAgrad, ADAM etc.) Convolutional Neural Networks (CNNs), Architectures few applications such as Object Localization Detection and Segmentation, Action Classification, sequential Models, BPTI, Problems with Recurrent neural Networks (RNNs), Long Short Term Memory (LSTM) / Gated Recurrent Units (GRUs), Bidirectional LSTMs (BLTSMs), Applications of RNNs to NLP, Language Model Unsupervised representational learning Auto Encoders and its variants, Variational inference and Deep learning: Variational Auto Encoders (VAE) and its variants, Generative Adversarial Networks Graph Convolutional networks (GCNs), Graph Attention Networks (GATs), Deep Reinforcement Learning, Advanced Topics: Fairness and Explainability in DL Adversarial Learning Neuro-Symbolic Reasoning, Neural Architecture Search.