Presenter: Noé Fouotsa Manfouo (PhD candidate in Operations Research and Machine Learning Fellow @ Fellowship AI)
Title of talk: Time series forecasting and machine learning
Summary:
The increasing availability of large amounts of historical data and the
need of performing accurate forecasting of future behaviour in several
scientific and applied domains demands the definition of robust and
efficient techniques able to infer from observations the stochastic
dependency between past and future. The forecasting domain has been
influenced, from the 1960s onward, by linear statistical methods
generally denoted as Box-Jenkins or autoregressive methods. These
methods have the main drawbacks of being limited to structured data
inferences, and not being able to predict extreme events. More recently,
machine learning models have drawn attention and have established
themselves as serious contenders to classical statistical models in the
forecasting community. This talk presents an overview of machine
learning techniques in time series forecasting by focusing on three
aspects: the formalization of one-step forecasting problems as
supervised learning tasks, the discussion of local learning techniques
as an effective tool for dealing with temporal data and the role of the
forecasting strategy when we move from one-step to multiple-step
forecasting. Snippets of python codes are also discussed for practical
implementation.
Keywords: Time series forecasting, machine learning, local learning, lazy learning, MIMO, python