Abstract: Techniques for generating distributed representations of computing processes and events are provided. According to one set of embodiments, a computer system can receive occurrence data pertaining to a plurality of computing processes and a plurality of events associated with the plurality of computing processes. The computer system can then generate, based on the occurrence data, (1) a set of distributed process representations that includes, for each computing process, a representation that encodes a sequence of events associated with the computing process in the occurrence data, and (2) a set of distributed event representations that includes, for each event, a representation that encodes one or more event properties associated with the event and one or more events that occur within a window of the event in the occurrence data.

Abstract: Techniques for implementing efficient federated learning of deep neural networks (DNNs) using approximation layers are provided. In one set of embodiments, given a DNN M with k original layers {L1, . . . , Lk}, k approximation layers {L1?, . . . , Lk?} can be created that correspond (i.e., map) to the k original layers. Each approximation layer can have the same number of inputs and outputs as its corresponding original layer, but can be smaller in size (i.e., have fewer parameters). Then, at the time of training DNN M via federated learning, for each participating client c during a training round r, a parameter server can transmit, for i=1, k, either (1) the current parameter values for approximation layer Li? alone, or (2) the current parameter values for both original layer Li and approximation layer Li? to client c. In response, client c can train its local copy of DNN M in accordance with the received parameter values.

Abstract: Techniques for generating distributed representations of computing processes and events are provided. According to one set of embodiments, a computer system can receive occurrence data pertaining to a plurality of computing processes and a plurality of events associated with the plurality of computing processes. The computer system can then generate, based on the occurrence data, (1) a set of distributed process representations that includes, for each computing process, a representation that encodes a sequence of events associated with the computing process in the occurrence data, and (2) a set of distributed event representations that includes, for each event, a representation that encodes one or more event properties associated with the event and one or more events that occur within a window of the event in the occurrence data.

Abstract: The disclosed computer-implemented method for training malware classifiers may include (1) perturbing, at a computing device, a binary file in a manner that maintains functionality of the binary file, (2) classifying the perturbed binary file with a first machine learning classifier to produce a classification result, (3) producing a transformed file by repeating the perturbing and classifying steps until the transformed file becomes misclassified, and (4) performing a security action comprising training a second machine learning classifier with the transformed file and an associated correct classification result. Various other methods, systems, and computer-readable media are also disclosed.

Abstract: The present disclosure relates to using correlations between support interaction data and telemetry data to discover emerging incidents for remediation. One example method generally includes receiving a corpus of support interaction data and a corpus of telemetry data. Topics indicative of underlying problems experienced by users of an application are extracted from the corpus of support interaction data. A topic having a rate of appearance in the support interaction data above a threshold value is identified. A set of telemetry data relevant to the topic is extracted from the corpus of telemetry data, and a subset of the relevant set of telemetry data having a frequency in the relevant set of telemetry data above a second threshold value is identified. The topic and the subset of telemetry data are correlated to an incident to be remediated, and one or more actions are taken to remedy the incident.

Abstract: The disclosed computer-implemented method for using electronic text information to automatically determine untrustworthy voice calls, at least a portion of the method being performed by a computing device comprising at least one processor, may include (1) during a voice call, receiving, by the computing device, text information representing contents of the voice call, (2) analyzing, by the computing device, the text information representing the contents of the voice call, (3) determining, by the computing device, that the voice call is untrustworthy based on the analysis of the text information, and (4) during the voice call, advising a recipient of the voice call of the determination that the voice call is untrustworthy. Various other methods, systems, and computer-readable media are also disclosed.