XLSTMTime: Long-Term Time Series Forecasting with xLSTM

Re:XLSTMTime: Long-Term Time Series Forecasting with xLSTM by techwizrd on Tuesday July 16, 2024 @02:27PM (#40978995)

In my aviation safety work, deep learning outperforms traditional non-DL models for multivariate time-series forecasting. Between deep learning models, I've had a wide variance in performance between transformers, Bi-LSTMs, regular MLPs, VAEs, and so on.

[ Parent | on HN ]

Re:XLSTMTime: Long-Term Time Series Forecasting with xLSTM by montereynack on Tuesday July 16, 2024 @03:43PM (#40979619)

Seconding the other question, would be curious to know

[ Parent | on HN ]

Re:XLSTMTime: Long-Term Time Series Forecasting with xLSTM by theLiminator on Tuesday July 16, 2024 @02:35PM (#40979055)

What's your go-to model that generally performs well with little tuning?

[ Parent | on HN ]

Re:XLSTMTime: Long-Term Time Series Forecasting with xLSTM by techwizrd on Tuesday July 16, 2024 @04:13PM (#40979886)

If you have short time-series with low variance, noise and outliers, strong prior knowledge, or limited resources to train and maintain a model, I would stick with simpler traditional models.

If DL is a good fit for your use-case, then I tend to like transformers or combining CNNs with recurrent models (e.g., BiGRU, GRU, BiLSTM, LSTM) and optional attention.

[ Parent | on HN ]

Re:XLSTMTime: Long-Term Time Series Forecasting with xLSTM by ramon156 on Tuesday July 16, 2024 @06:18PM (#40980842)

Now take into account that it has to be lightweight and DL falls shirt

[ Parent | on HN ]

Re:XLSTMTime: Long-Term Time Series Forecasting with xLSTM by nerdponx on Tuesday July 16, 2024 @06:40PM (#40980991)

What are you doing in aviation safety that requires time series modeling? Weather?

[ Parent | on HN ]

Re:XLSTMTime: Long-Term Time Series Forecasting with xLSTM by all2 on Wednesday July 17, 2024 @12:54AM (#40982679)

My best guess would be accident occurrence prediction.

[ Parent | on HN ]