Abstract: An apparatus to facilitate neural network (NN) training is disclosed. The apparatus includes training logic to receive one or more network constraints and train the NN by automatically determining a best network layout and parameters based on the network constraints.

Abstract: An apparatus to facilitate optimization of a neural network (NN) is disclosed. The apparatus includes optimization logic to define a NN topology having one or more macro layers, adjust the one or more macro layers to adapt to input and output components of the NN and train the NN based on the one or more macro layers.

Abstract: A mechanism is described for facilitating recognition, reidentification, and security in machine learning at autonomous machines. A method of embodiments, as described herein, includes facilitating a camera to detect one or more objects within a physical vicinity, the one or more objects including a person, and the physical vicinity including a house, where detecting includes capturing one or more images of one or more portions of a body of the person. The method may further include extracting body features based on the one or more portions of the body, comparing the extracted body features with feature vectors stored at a database, and building a classification model based on the extracted body features over a period of time to facilitate recognition or reidentification of the person independent of facial recognition of the person.

Abstract: A mechanism is described for facilitating sharing of data and compression expansion of models at autonomous machines. A method of embodiments, as described herein, includes detecting a first processor processing information relating to a neural network at a first computing device, where the first processor comprises a first graphics processor and the first computing device comprises a first autonomous machine. The method further includes facilitating the first processor to store one or more portions of the information in a library at a database, where the one or more portions are accessible to a second processor of a computing device.

Abstract: In an example, an apparatus comprises a compute engine comprising a high precision component and a low precision component; and logic, at least partially including hardware logic, to receive instructions in the compute engine; select at least one of the high precision component or the low precision component to execute the instructions; and apply a gate to at least one of the high precision component or the low precision component to execute the instructions. Other embodiments are also disclosed and claimed.

Abstract: In an example, an apparatus comprises a plurality of execution units comprising at least a first type of execution unit and a second type of execution unit and logic, at least partially including hardware logic, to analyze a workload and assign the workload to one of the first type of execution unit or the second type of execution unit. Other embodiments are also disclosed and claimed.

Abstract: A mechanism is described for facilitating barriers and synchronization for machine learning at autonomous machines. A method of embodiments, as described herein, includes detecting thread groups relating to machine learning associated with one or more processing devices. The method may further include facilitating barrier synchronization of the thread groups across multiple dies such that each thread in a thread group is scheduled across a set of compute elements associated with the multiple dies, where each die represents a processing device of the one or more processing devices, the processing device including a graphics processor.

Abstract: An apparatus to facilitate compute optimization is disclosed. The apparatus includes sorting logic to sort processing threads into thread groups based on bit depth of floating point thread operations.

Abstract: An apparatus to facilitate compute optimization is disclosed. The apparatus includes sorting logic to sort processing threads into thread groups based on bit depth of floating point thread operations.

Abstract: A mechanism is described for facilitating recognition, reidentification, and security in machine learning at autonomous machines. A method of embodiments, as described herein, includes facilitating a camera to detect one or more objects within a physical vicinity, the one or more objects including a person, and the physical vicinity including a house, where detecting includes capturing one or more images of one or more portions of a body of the person. The method may further include extracting body features based on the one or more portions of the body, comparing the extracted body features with feature vectors stored at a database, and building a classification model based on the extracted body features over a period of time to facilitate recognition or reidentification of the person independent of facial recognition of the person.

Abstract: In an example, an apparatus comprises a plurality of processing unit cores, a plurality of cache memory modules associated with the plurality of processing unit cores, and a machine learning model communicatively coupled to the plurality of processing unit cores, wherein the plurality of cache memory modules share cache coherency data with the machine learning model. Other embodiments are also disclosed and claimed.

Abstract: Methods and apparatus relating to autonomous vehicle neural network optimization techniques are described. In an embodiment, the difference between a first training dataset to be used for a neural network and a second training dataset to be used for the neural network is detected. The second training dataset is authenticated in response to the detection of the difference. The neural network is used to assist in an autonomous vehicle/driving. Other embodiments are also disclosed and claimed.

Abstract: Systems and methods may provide for determining a usage configuration of a wearable device and setting an activation state of an air conduction speaker of the wearable device based at least in part on the usage configuration. Additionally, an activation state of a tissue conduction speaker of the wearable device may be set based at least in part on the usage configuration. In one example, the usage configuration is determined based on a set of status signals that indicate one or more of a physical position, a physical activity, a current activation state, an interpersonal proximity state or a manual user request associated with one or more of the air conduction speaker or the tissue conduction speaker.

Abstract: One or more sensors gather data, one or more processors analyze the data, and one or more indicators notify a user if the data represent an event that requires a response. One or more of the sensors and/or the indicators is a wearable device for wireless communication. Optionally, other components may be vehicle-mounted or deployed on-site. The components form an ad-hoc network enabling users to keep track of each other in challenging environments where traditional communication may be impossible, unreliable, or inadvisable. The sensors, processors, and indicators may be linked and activated manually or they may be linked and activated automatically when they come within a threshold proximity or when a user does a triggering action, such as exiting a vehicle. The processors distinguish extremely urgent events requiring an immediate response from less-urgent events that can wait longer for response, routing and timing the responses accordingly.

Abstract: Technologies for presenting an advertisement on a media consumption device includes receiving a request to seek past a commercial included in media content played on the media consumption device, determining an advertisement based on the commercial, and presenting the advertisement to a user of the media consumption device during performance of the requested seek function. The advertisement may be, for example, an extracted frame or image of the commercial and may include a logo or phrase associated with a product or service advertised in the commercial. Similar technologies related to a media content distribution system are also disclosed.

Abstract: Technologies for bio-chemically controlling operation of a machine include applying a bio-chemical agent to an operator of the machine and controlling an operational characteristic of the machine based on the presence of the bio-chemical agent on the operator. The operational characteristic of the machine may be controlled based on the presence or lack of the biochemical agent on the operator. In some embodiments, the bio-chemical agent may be configured to generate a bio-chemical trigger in response to exposure to a biochemical or biological characteristic of the operator. The operation of the machine may be controlled based on such bio-chemical trigger or reaction.

Abstract: Techniques to project an image from a wearable computing device are provided. A wearable computing device including a projector configured to project an image into a user field of view based on output from one or more sensors and/or images captured by a camera. The wearable computing device can also include a touch input device. The wearable computing device can project an image responsive to a users touch based on signals received from the touch input device.

Abstract: Various systems and methods for improving map and navigation data are described herein. An electronic navigation system for improving map and navigation data comprises a database access module to access a database of physiological information to obtain a biometric value, the biometric value associated with a location and a time; a processing module to determine whether the biometric value violates a threshold; and a display module to display a notification on a map when the threshold is violated, the map including an area around the location associated with the biometric value, and the notification displayed proximate to the location associated with the biometric value.

Abstract: Technologies for managing network privileges of members of graft-network include detecting a computing device in physical presence with a network infrastructure, determining whether the computing device is a member of the graft-network, and establishing initial network privileges for the computing device if the computing device is not a member, without direct programming of the member. The network privileges of members of the graft-network are updated over time as a function of the length of time for which the computing device is in physical presence of the network infrastructure. A computing device may be in physical presence of the network by physical contacting a communication bus of the network infrastructure or being within a limited communication range of the communication bus. New members to the graft-network may be quarantined to reduce risk to the network.

Abstract: Technologies for managing security threats on a computing system include detecting a security threat to the computing system, determining a plurality of mitigation scenarios to employ on the computing system to mitigate the security threat, and implementing the plurality of mitigation scenarios. Each mitigation scenario includes one or more threat mitigation actions to be taken by the computing system, one or more response systems of the computing system to perform the threat mitigation actions, and a temporal sequence in which the threat mitigation actions are to be taken. The results of each mitigation scenario is evaluated and a validated mitigation scenario is determined based on the results. A user of the computing device may be subsequently trained or habituated to mitigate the security threat by requesting interaction from the user during the implementation of the validated mitigation scenario in response to a threat scenario designed to replicate the security threat.