Upgrade from LSTMs

Solve time series problems more accurately and efficiently with LMUs

Long short-term memories (LSTMs) have been called the most commercially valuable AI algorithm ever invented (Bloomberg), having been deployed in many speech, vision and text analysis systems. But LSTMs fall apart when tasked with learning temporal dependencies spanning thousands of timesteps, limiting their usefulness.

We have developed a new algorithm called the Legendre Memory Unit (LMU) that can learn temporal dependencies over millions of timesteps. LMUs maintain efficient and scale-invariant representations of recent inputs and learn how to solve real world problems using those representations. LMUs can be implemented with traditional deep learning techniques on hardware you already have, or can be deployed on neuromorphic hardware for massive power savings. Using LMUs, we have obtained the best-known results on the permuted sequential MNIST task, a difficult RNN benchmark.

Do you have a time series problem you’re currently solving with an LSTM?

Have a time series problem that you haven’t been able to solve yet?

Do you need to solve a time series problem in an edge or IoT device?

Take a look at our published and easily reproducible results. ABR’s patent-pending LMU is free to use for academic research and personal non-commercial uses. For commercial uses, please contact us through the form below.