Abstract: An artificial neural network (ANN) that learns at the Edge (e.g., on a smart phone) can be faster and use less network bandwidth than an ANN trained on a server and distributed to the Edge. Learning at the compute edge can be accomplished by executing Lifelong Deep Neural Network (L-DNN) technology at the compute edge. L-DNN technology uses a representation-rich, DNN-based subsystem with a fast-learning subsystem to learn new features quickly without forgetting previously learned features. Compared to a conventional DNN, L-DNN uses much less data to build robust networks, has dramatically shorter training time, and learns on-device instead of on servers without re-training or storing data. An edge device with L-DNN can learn continuously after deployment, eliminating costs in data collection and annotation, memory, and compute power. This fast, local, on-device learning can be used in unsupervised mode to make personal assistants more intelligent and enhance frequently used apps.

Abstract: A system for assisting a user in selecting a packet-based telephony service provider (“PBTSP”) from a group of PBTSPs, and in registering the user with the selected PBTSP for outbound and/or inbound calling. The selection of a particular PBTSP may be stored in a database, and this stored information may be used to enable the user to exchange telephone call data with one of the PBTSPs. One preferred system receives data from the user's computer device, determines whether the user is registered with one the PBTSPs and, if so, transmits data to the computer device which enables it to exchange telephone call data with one of the PBTSPs. The system also receives telephone call data from the computer device, accesses contact information for a PBTSP, and transmits the received telephone call data to the PBTSP using the accessed contact information.

Abstract: A system for monitoring user activity in an instant messaging session on a computer network periodically sends an activity message to other participants in the instant messaging session if the user has actively entered data during a first predetermined time interval. The system periodically sends a new activity message at intervals corresponding to the first predetermined time interval so long as the user is actively entering data during each time interval. If the user has not entered data during the first predetermined time interval, the system will not send an activity message. Other participants in the instant messaging session receive the activity message and generate an activity indicator on their respective displays. The computer receiving the activity message displays an activity indicator on the computer display in response to receipt of an activity message and starts a timer to measure a second predetermined time interval.

Abstract: A system for monitoring user activity in an instant messaging session on a computer network periodically sends an activity message to other participants in the instant messaging session if the user has actively entered data during a first predetermined time interval. The system periodically sends a new activity message at intervals corresponding to the first predetermined time interval so long as the user is actively entering data during each time interval. If the user has not entered data during the first predetermined time interval, the system will not send an activity message. Other participants in the instant messaging session receive the activity message and generate an activity indicator on their respective displays. The computer receiving the activity message displays an activity indicator on the computer display in response to receipt of an activity message and starts a timer to measure a second predetermined time interval.