Abstract: In an example of the disclosed techniques, a method comprising computing, by a computing system, a distribution of symbol errors in a forward error correction (FEC) frame transmitted in an optical signal. The method further includes, in response to determining that the distribution of symbol errors indicates a burst error that satisfies a threshold, adjusting, in an optical transceiver module, one or more of a pre-cursor setting, a post-cursor setting, or a signal amplitude for the optical signal.

Abstract: An example cascaded optical switch includes: a fast optical switch and a slow optical switch both having input ports and output ports. A switching time of the slow switch is longer than a switching time of the fast optical switch. The cascaded optical switch includes a first pre-cabled optical fiber connecting an output port of the slow optical switch to an input ports of the fast optical switch, and a second pre-cabled optical fiber connecting an output ports of the fast optical switch to an input ports of the slow optical switch. The fast optical switch or the slow optical switch is configured to receive configuration data to modify a switching configuration to configure a fiber cross-connect in the optical switch fabric that includes at least one of the first pre-cabled optical fiber or the second pre-cabled optical fiber.

Abstract: An example optical switch includes a plurality of input ports and a plurality of output ports, a cross-connect fabric having one or more inputs, one or more outputs, and a device to selectively cross-connect the inputs with the outputs. The optical switch includes an integrated fast optical switch comprising a first input, a first output, and a second output, wherein the first input is connected to a first one of the outputs of the cross-connect fabric, and wherein the integrated fast optical switch has a switching time that is less than a switching time of the cross-connect fabric to switch the first input between the first output on a path to a first output port of the plurality of output ports and the second output on a path to a second output port of the plurality of output ports.

Abstract: Techniques are described for providing a hybrid compensation of chromatic dispersion in optical networks to reduce power consumption by coherent receivers. In some examples, a controller may receive a chromatic dispersion value of an optical signal from a coherent receiver integrated with a receiver optical network device. The controller may compare the chromatic dispersion value with a threshold. The controller may, in response to determining that the chromatic dispersion value satisfies the threshold, perform at least one of: configure a switch connected to a dispersion compensation module (DCM) with a state to provide access to the DCM to compensate the chromatic dispersion value of the optical signal, or adjust a phase response of a filter of a coherent transmitter to compensate the chromatic dispersion value of the optical signal.

Abstract: Techniques for machine learning using curated features are provided. A plurality of key terms is identified for a first document type of a plurality of document types. A document associated with the first document type is received, and the document is modified by inserting one or more of the plurality of key terms. A vector is generated for the modified document, and a machine learning model is trained to categorize input into the plurality of document types based on the modified document.

Abstract: Techniques are provided for delayed processing for arm policy determination for content management system messaging, including, during a delayed processing window, receiving reward data for arm actions taken, where the arm actions were chosen based on a previous version of an arm choice policy, and the previous version of the arm choice policy was determined based on a previous set of reward data for a previous set of arm actions taken. When the delayed processing window has closed, a new arm choice policy is determined based at least in part on the action-reward data, and the previous set of reward data and/or the previous arm choice policy. After a request to choose an arm choice is received, a particular arm action to take is determined based on the new arm choice policy. This chosen arm is provided in response to the request.

Abstract: Techniques for event-based electronic communications management. In one embodiment, for example, the techniques include obtaining a directed graph representing a messaging track. Based on detecting an event pertaining to a user, the directed graph is traversed via a directed edge of the graph from a non-messaging action node of the graph to a messaging action node of the graph. Based on traversing the directed graph, the messaging action is caused to be performed. The techniques enable more targeted and timely event-driven electronic messaging campaigns.

Abstract: A real-time trigger for event-based electronic communication system messaging. In one embodiment, for example, an event caused by a user's activity or inactivity with respect to an online service is queued. When the event is dequeued, it is transformed to a trigger and the trigger is mapped to an action. The action is dispatched to a messaging track actor to carry out the action. Carrying out the action may include putting a user on a messaging track or transitioning the state of a user already on a messaging track.

Abstract: Techniques are provided for delayed processing for arm policy determination for content management system messaging, including, during a delayed processing window, receiving reward data for arm actions taken, where the arm actions were chosen based on a previous version of an arm choice policy, and the previous version of the arm choice policy was determined based on a previous set of reward data for a previous set of arm actions taken. When the delayed processing window has closed, a new arm choice policy is determined based at least in part on the action-reward data, and the previous set of reward data and/or the previous arm choice policy. After a request to choose an arm choice is received, a particular arm action to take is determined based on the new arm choice policy. This chosen arm is provided in response to the request.

Abstract: Computer-implemented techniques include, during a delayed processing window, receiving reward data for arm actions taken, where the arm actions were chosen based on a previous version of an arm choice policy, and the previous version of the arm choice policy was determined based on a previous set of reward data for a previous set of arm actions taken. When the delayed processing window has closed, a new arm choice policy is determined based at least in part on the action-reward data, and the previous set of reward data and/or the previous arm choice policy. After a request to choose an arm choice is received, a particular arm action to take is determined based on the new arm choice policy. This chosen arm is provided in response to the request.

Abstract: A real-time trigger for event-based electronic communication system messaging. In one embodiment, for example, an event caused by a user's activity or inactivity with respect to an online service is queued. When the event is dequeued, it is transformed to a trigger and the trigger is mapped to an action. The action is dispatched to a messaging track actor to carry out the action. Carrying out the action may include putting a user on a messaging track or transitioning the state of a user already on a messaging track.

Abstract: Techniques for event-based electronic communications management. In one embodiment, for example, the techniques include obtaining a directed graph representing a messaging track. Based on detecting an event pertaining to a user, the directed graph is traversed via a directed edge of the graph from a non-messaging action node of the graph to a messaging action node of the graph. Based on traversing the directed graph, the messaging action is caused to be performed. The techniques enable more targeted and timely event-driven electronic messaging campaigns.

Abstract: Techniques are provided for delayed processing for arm policy determination for content management system messaging, including, during a delayed processing window, receiving reward data for arm actions taken, where the arm actions were chosen based on a previous version of an arm choice policy, and the previous version of the arm choice policy was determined based on a previous set of reward data for a previous set of arm actions taken. When the delayed processing window has closed, a new arm choice policy is determined based at least in part on the action-reward data, and the previous set of reward data and/or the previous arm choice policy. After a request to choose an arm choice is received, a particular arm action to take is determined based on the new arm choice policy. This chosen arm is provided in response to the request.

Abstract: Described is transparently compressing content for network transmission, including end-to-end compression. An end host or middlebox device sender sends compressed packets to an end host or middlebox device receiver, which decompresses the packets to recover the original packet. The sender constructs compressed packets including references to information maintained at the receiver, which the receiver uses to access the information to recreate actual original packet content. The receiver may include a dictionary corresponding to the sender, e.g., synchronized with the sender's dictionary. Alternatively, in speculative compression, the sender does not maintain a dictionary, and instead sends a fingerprint (hash value) by which the receiver looks up corresponding content in its dictionary; if not found, the receiver requests actual content.

Abstract: Described is transparently compressing content for network transmission, including end-to-end compression. An end host or middlebox device sender sends compressed packets to an end host or middlebox device receiver, which decompresses the packets to recover the original packet. The sender constructs compressed packets including references to information maintained at the receiver, which the receiver uses to access the information to recreate actual original packet content. The receiver may include a dictionary corresponding to the sender, e.g., synchronized with the sender's dictionary. Alternatively, in speculative compression, the sender does not maintain a dictionary, and instead sends a fingerprint (hash value) by which the receiver looks up corresponding content in its dictionary; if not found, the receiver requests actual content.