This publication describes methods for learning a user’s intent, when the user interacts with multiple computing devices, based on detected spatial gestures. The methods may utilize a machine-learned model with a reinforcement-learning framework trained to classify spatial gestures, user state information, and/or device state information detected during a device-to-device interaction, generate predictions relating to the intent of the user, and utilize the predictions to facilitate the device-to-device interaction.

Creative Commons License

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.