Multiple Instance Learning for Efficient Sequential Data Classification on Resource-constrained Devices

This enables us to deploy such models for continuous real-time prediction on devices as small as a Raspberry Pi0 and Arduino variants, a task that the baseline LSTM could not achieve.