Mediaspace scheduled maintenance: Aug 25, 2026 07:00 - 12:00 AM. During this time, videos will be temporarily unavailable. Check status updates.

Home

Course

CS-456: Deep reinforcement learning

About
Privacy
Disclaimer

Graph Chatbot

Lectures in this course (96)

Interpreting Output as Probability

Delves into interpreting neural network output as probabilities based on the cross-entropy error function.

Deep Q-Learning: DeepRL1.1

Covers Deep Q-learning in deep neural networks, its application in games, backpropagation, Q-values, and V-values.

Statistical Interpretation of Artificial Neural Networks

Delves into the statistical interpretation of artificial neural networks, exploring the likelihood of data and maximizing model accuracy.

Sigmoidal Units: Natural Output Functions

Delves into using sigmoidal units as natural output functions in deep learning, focusing on statistical interpretation and optimal derivation.

Multi-class Problems: One-hot Coding

Explores one-hot coding for exclusive classes and the cross-entropy loss function in multi-class problems.

Statistical Approach: Summary and Quiz

Discusses statistical view of neural networks, classification tasks, and cross-entropy loss functions.

Inductive Bias in Machine Learning

Explores the concept of inductive bias in machine learning, emphasizing the role of prior knowledge in designing effective neural networks.

Convolutional filters as inductive bias for images

Delves into convolutional filters as an inductive bias for images in neural networks, emphasizing independence to translation and local feature detectors.

MaxPooling as inductive bias for images

Explores how MaxPooling enforces an inductive bias towards local translation invariance in convolutional neural networks.

Convolutional Layers: Quiz

Presents a quiz on convolutional layers, exploring weights, invariance, and pooling methods.

Convolutional Layer: The Gradient

Explores the optimization of filters in a convolutional layer and the backpropagation process for MaxPooling.

Automatic Differentiation: BackProp revisited

Discusses automatic differentiation, emphasizing reverse mode differentiation for optimizing convolutional layer filters by gradient descent.

Gradient Descent Methods for Artificial Neural Networks

Explores gradient descent methods for training artificial neural networks, covering supervised learning, single-layer networks, and modern gradient descent rules.

Reducing the number of parameters: Outer-product representation

Covers reducing parameters in convolutional filters through outer-product representation.

Modern Convolutional Networks and Image Recognition

Explores the evolution of deep convolutional networks and their impact on image recognition accuracy.

Convolutional Networks: Applications beyond Object Recognition

Delves into the applications of convolutional networks beyond object recognition, emphasizing their impact on neuroscience, brain sciences, and art.

Page 5 of 5