A model requester node, which is an edge node of a cloud computing network, generates a specification of a machine learning model, distributes the specification to a plurality of other edge nodes, and receives replies to the specification from the plurality of other edge nodes. In response to the replies, the model requester node identifies a set of participating edge nodes based on a learning utility and a cost estimate of each of the plurality of other edge nodes. The model requester node then trains the machine learning model, without exchanging training data among the model requester node and the participating edge nodes, by repeatedly: distributing most recent parameters of the machine learning model to the participating edge nodes; receiving updates to the most recent parameters from the participating edge nodes; and establishing new parameters for the machine learning model by aggregating the updates from the participating edge nodes.

A method for distributed neural network processing, the method may include detecting, by a local neural network that belongs to a local device, and based on sensed information, an occurrence of a triggering event for executing or completing a classification or detection process; sending to a remote device, a request for executing or completing the classification or detection process by a remote device that comprises a remote neural network; wherein the remote neural network has more computational resources than the local neural network; determining by the remote device whether to accept the request; and executing or completing, by the remote device, the classification or detection process when determining to accept the request; wherein the executing or completing involves utilizing the remote neural network.

A computer-implemented method, a computer program product, and a computer system for enhanced distributed machine learning. A fusion server in a distributed machine learning system determines correlation relationships across agents in the distributed machine learning system, based on auxiliary information. The fusion server clusters the agents to form one or more communities, based on the correlation relationships. The fusion server selects, from the one or more communities, participating agents that participate in the enhanced distributed machine learning.

An approach is provided in which the approach applies, by a first node, a first axiom to a set of data points to generate a set of first outputs. The approach applies, by a second node, a second axiom to the set of data points to generate a set of second outputs. The first node and the second node are part of a computer network that includes multiple nodes. The approach computes a first nuance based on a set of disagreements between the set of first outputs and the set of second outputs, and adjusts a reliability of the first node in the computer network based on the first nuance.

A computer-implemented method includes determining, by a master node, model update information at least based on a workload related to a task and a resource capacity of a computing environment. The model update information indicates respective model update suggestions for a plurality of inference models configured to perform the task. The method further includes distributing, by the master node, the model update information to a plurality of inference agents in the computing environment. The plurality of inference agents has a plurality of instances of the plurality of inference models executed thereon.

Techniques for agent-assist systems to provide context-aware, subdocument-granularity recommended answers to agents that are attempting to answer queries of users. The agent-assist system may obtain collections of documents that include information for responding to queries, and analyze those documents to identify subdocuments that are associated with different semantics or meanings. Subsequently, any queries received can be analyzed to identify their semantics, and relevant subdocuments can be identified as having similar semantics. When the agent-assist system presents the agent with the relevant documents, it may highlight or otherwise indicate the relevant subdocument within the document for quick identification by the agent. Further, the agent-assist system may collect feedback from the agent and/or user to determine a relevancy of the recommended answers. The agent-assist system can use the feedback to improve the quality of the recommended answers provided to the agents.

An information processing system according to an embodiment includes edge terminals, an information processing apparatus, and one or more service providing apparatuses. The edge terminals each transmit, to the information processing apparatus, monitoring data indicating a state of a target system to be analyzed. The information processing apparatus performs an analysis process by using an analysis mod& that inputs an input value including the monitoring data transmitted from the edge terminals and outputs an output value of a value function. The analysis process is a process to obtain the output value in response to the input value, The information processing apparatus transmits, to the service providing apparatuses, information indicating an analysis result of the analysis process. The service providing apparatuses each output information obtained by visualizing the analysis result on the basis of information indicating the analysis result.

Cloud infrastructure planning systems and methods can utilize artificial intelligence/machine learning agents for developing a plan of demand, plan of record, plan of execution, and plan of availability for developing cloud infrastructure plans that are more precise and accurate, and that learn from previous planning and deployments. Some agents include one or more of supervised, unsupervised, and reinforcement machine learning to develop accurate predictions and perform self-tuning alone or in conjunction with other agents.

The bionic computing system includes a perception subsystem, an attention subsystem, and a temporal-spatial awareness subsystem. The perception subsystem has several perceptual devices for detecting objects from sequences of sensory data and generating an object record for each object. The attention subsystem adjusts the object records by re-identifying the tracking identities across sensory devices, generates several object associations, generate several location associations, and generates several motion implications. The temporal-spatial awareness subsystem organizes and retains the object records in a working memory space. The perception subsystem identifies several basic events from each sensory datum of the same perceptual device, the attention subsystem identifies an episodic event by determining that a portion of the basic events from the same sensory device within a temporal window conform to a pattern, and the temporal-spatial awareness subsystem identifies a complex event according the detected episodic event and the object records in the working memory space.

There is disclosed a method and system for operating an event-driven architecture. The event-driven architecture comprises a first machine-learning (ML) agent operating a first service and a second ML agent operating a second service. The first ML agent comprises a first model and first model metadata. The second ML agent comprises a second model and second model metadata. The method comprises generating, by the first ML agent, an event associated with event metadata. The event comprises results generated by the first model. The event metadata comprises an event identifier (ID). The first ML agent publishes the event in a virtualized dedicated space. The second ML agent determines whether the event is to be processed by the second ML agent. If a determination is made that the message is to be processed by the second ML agent, the second ML agent processes the event to generate an output.