Network Architectures - Part 1
-
- 0 Rating
- 0 Reviews
- 0 Students Enrolled
Network Architectures - Part 1
Sequential modeling is framed from simple time-series baselines to neural generalizations, showing how networks capture nonlinear and long-range patterns. Language models are defined as probabilistic sequence predictors, and RNNs—via recurrent cells and their unrolled computational graphs—are illustrated with practical language-modeling examples.
-
- 0 Rating
- 0 Reviews
- 0 Students Enrolled
Requirements
- see DELTA course requirements
General Overview
Description
This unit introduces network architectures for sequential data through a practical time-series forecasting primer: starting from simple baselines (last value, moving average) and extending to generalized setups that account for trend, seasonality, exogenous inputs, and multi-step horizons—highlighting where neural networks offer advantages for nonlinear and long-range dependencies. It then formalizes language models as systems that assign probabilities to sequences for prediction and generation, emphasizing probabilistic forecasting, typical requirements (tokenization choices, adequate context windows, sizable corpora, and sound evaluation such as perplexity), and constraints. The core machinery, Recurrent Neural Networks (RNNs), is presented via recurrent cells (vanilla, LSTM, GRU), their computational graph when unrolled through time, and concrete language-modeling examples (character- and word-level next-token prediction), linking training dynamics to sequence understanding.
Courses that include this CL
Recommended for you
Meet the instructors !
Stefan received a PhD from the University of Hamburg in 2007, where he also completed his habilitation on decision analysis and support using ensemble forecasting models in 2012. He then joined the Humboldt-University of Berlin in 2014, where he heads the Chair of Information Systems at the School of Business and Economics. He serves as an associate editor for the International Journal of Business Analytics, Digital Finance, and the International Journal of Forecasting, and as department editor of Business and Information System Engineering (BISE). Stefan has secured substantial amounts of research funding and published several papers in leading international journals and conferences. His research concerns the support of managerial decision-making using quantitative empirical methods. He specializes in applications of (deep) machine learning techniques in the broad scope of marketing and risk analytics. Stefan actively participates in knowledge transfer and consulting projects with industry partners; from start-up companies to global players and not-for-profit organizations.