Abstract: A system can be provided for deploying bare metal clusters that satisfy custom resource requests. For example, the system can receive from a client device, a custom resource request. The custom resource request can include a set of requirements for a bare metal cluster. The set of requirements can include a number of nodes for the bare metal cluster. The system can determine a set of resources that satisfies the set of requirements. The set of resources can include virtual Internet Protocol (IP) addresses and a set of baseband management controller (BMC) IP addresses. A number of BMC IP addresses in the set of BMC IP addresses can be equal to the number of nodes for the bare metal cluster. Additionally, the system can generate, based on the set of resources, a configuration file for the bare metal cluster and deploy, based on the configuration file, the bare metal cluster.

Abstract: In one embodiment, an apparatus includes: a plurality of execution circuits to execute and instruct micro-operations (?ops), where a subset of the plurality of execution circuits are capable of execution of a fused ?op; a fusion circuit coupled to at least the subset of the plurality of execution circuits, wherein the fusion circuit is to fuse at least some pairs of producer-consumer ?ops into fused ?ops; and a fusion throttle circuit coupled to the fusion circuit, wherein the fusion throttle circuit is to prevent a first ?op from being fused with another ?op based at least in part on historical information associated with the first ?op. Other embodiments are described and claimed.

Abstract: Job execution environment control techniques are described to manage policy selection and implementation to control use of job executors by a computing device, automatically and without user intervention. These techniques are usable to select a policy from a plurality of policies that is then used to control lifecycles of job executors of a job execution environment of a computing device. Further, these techniques are usable to respond dynamically to change the selected policy during runtime of the application in response to changes in the job execution environment.

Abstract: A signal processing device and a display apparatus for vehicles including the same are disclosed. The signal processing device includes a processor configured to perform signal processing for a display located in a vehicle, wherein the processor is configured to execute first to third virtual machines on a hypervisor in the processor, the second virtual machine is operated for a first display, the third virtual machine is operated for a second display, and the first virtual machine in the processor is configured to share at least some of data with the second virtual machine and the third virtual machine for processing of divided data. Consequently, the plurality of virtual machines for the plurality of displays in the vehicle may divide and process data.

Abstract: A method, computer program product and computer system for predicting excessive resource usage in a distributed computing environment is provided. A processor retrieves a portion of code associated with a workload in a distributed computing environment. A processor retrieves account information associated with the workload. A processor determines a likelihood that the workload is indicative of excessive resource usage based on the portion of code and the account information associated with the workload. In response to the likelihood of excessive resource usage exceeding a threshold, a processor reschedules the workload in the distributed computing environment.

Abstract: A computer-implemented method of monitoring programmatic containers performed through executing an agent processor is disclosed. The method comprises transmitting, by a processor, one or more deployment configurations from a monitoring server related to an application hosted in a container to a backend device, the processor receiving, from the backend device, a plurality of monitoring configurations for the application, the processor merging the plurality of monitoring configurations for the application into a merged monitoring configuration for the application, the processor providing the merged monitoring configuration for the application to the monitoring server, and the processor periodically receiving, from the monitoring server, telemetry data that characterizes one or more instances of the application.

Abstract: Systems and methods are provided for lock-free thread scheduling. Threads may be placed in a ring buffer shared by all computer processing units (CPUs), e.g., in a node. A thread assigned to a CPU may be placed in the CPU's local run queue. However, when a CPU's local run queue is cleared, that CPU checks the shared ring buffer to determine if any threads are waiting to run on that CPU, and if so, the CPU pulls a batch of threads related to that ready-to-run thread to execute. If not, an idle CPU randomly selects another CPU to steak threads from, and the idle CPU attempts to dequeue a thread batch associated with the CPU from the shared ring buffer. Polling may be handled through the use of a shared poller array to dynamically distribute polling across multiple CPUs.

Abstract: Example embodiments relate generally to systems and methods for continuous data protection (CDP) and more specifically to an input and output (I/O) filtering framework and log management system to seek a near-zero recovery point objective (RPO).

Abstract: Application migration using data movement capabilities of a storage system, including: executing an application on a first on-premises cloud infrastructure; storing, in a first storage system that is coupled for data communications with the first on-premises cloud infrastructure, data associated with the application; replicating, from the first storage system to a second storage system that is coupled for data communications with a second on-premises cloud infrastructure, the data; and initiating, based on the data, execution of a new instance of the application on the second on-premises cloud infrastructure.

Abstract: Techniques are disclosed relating to training a machine learning model to understand one or more rules without explicitly executing the rule. In some embodiments, a computer system generates synthetic samples for a trained machine learning model usable to make a classification decision, where the synthetic samples are generated from a rule and a set of existing samples. In some embodiments, the set of existing samples are selected based on exceeding a confidence threshold for the classification decision. In some embodiments, the computer system retrains the trained machine learning model using the synthetic samples.

Abstract: A method for the detection of anomalies in a networked water distribution system is provided. It improves detection methods based on an iterative modification of control variables of the network, by determining a reduced set of entities of the water distribution network on which control variables should be iteratively modifier. The invention increases the computing costs, and the reliability of such methods of detecting anomalies in a water distribution system.

Abstract: A framework for interpreting machine learning models is proposed that utilizes interpretability methods to determine the contribution of groups of input variables to the output of the model. Input variables are grouped based on dependencies with other input variables. The groups are identified by processing a training data set with a clustering algorithm. Once the groups of input variables are defined, scores related to each group of input variables for a given instance of the input vector processed by the model are calculated according to one or more algorithms. The algorithms can utilize group Partial Dependence Plot (PDP) values, Shapley Additive Explanations (SHAP) values, and Banzhaf values, and their extensions among others, and a score for each group can be calculated for a given instance of an input vector per group. These scores can then be sorted, ranked, and then combined into one hybrid ranking.

Abstract: Systems and methods are provided for reusing machine learning models. For example, the applicability of prior models may be compared using one or more assessment values, including a similarity threshold and/or an accuracy threshold. The similarity threshold may identify a similarity of data between a first data set used to generate a first model and a new data set that is received by the system. When the similarity between these two data sets is exceeded, the system may reuse a model with the highest similarity value. When an accuracy value of the data set does not exceed an accuracy threshold, the system may initiate a retraining process to generate a second ML model associated with the second data.

Abstract: There is described methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training a reinforcement learning system. The reinforcement learning system comprises an agent configured to perform actions based upon a policy and an intrinsic reward system configured to generate intrinsic reward values for the agent based upon the actions taken by the agent. The method comprises training the reinforcement learning system based upon a plurality of tasks. The training comprises updating the agent's policy based upon the intrinsic reward values generated by the intrinsic reward system and updating the intrinsic reward system based upon an extrinsic reward value obtained based upon the task being performed by the agent. The training further comprises re-initializing the agent's policy when an expiration criterion associated with the agent is met.

Abstract: A system uses machine learning models, such as neural networks for generating mask design from a circuit design. The machine learning models have inputs and outputs which are localized to a small region of the circuit design. The machine learning model takes as input features describing the circuit design in the neighborhood of a location and generates an offset distance as output. The system uses the offset distance to generate features of the mask design, for example, main features or assist features corresponding to a circuit design polygon. The system may use the offset distance for target optimization by modifying the circuit design polygon to obtain a circuit design polygon that has improved manufacturability.

Abstract: An artificial intelligence (AI) system includes an agent to learn a policy and provide an action; the agent can be a neural network. The AI system further includes a processor to process information associated with the action and provide a state and a reward to the agent. The state is based on a number of state variables, and the agent further updates the policy based on multiple updates of the state variables to achieve the highest reward.

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: In various examples, a three-dimensional (3D) intersection structure may be predicted using a deep neural network (DNN) based on processing two-dimensional (2D) input data. To train the DNN to accurately predict 3D intersection structures from 2D inputs, the DNN may be trained using a first loss function that compares 3D outputs of the DNN—after conversion to 2D space—to 2D ground truth data and a second loss function that analyzes the 3D predictions of the DNN in view of one or more geometric constraints—e.g., geometric knowledge of intersections may be used to penalize predictions of the DNN that do not align with known intersection and/or road structure geometries. As such, live perception of an autonomous or semi-autonomous vehicle may be used by the DNN to detect 3D locations of intersection structures from 2D inputs.

Abstract: SR-IOV (single root IO virtualization) capable PCIe devices can implement virtual functions (VFs) that are assigned to VMs running on a host machine, thereby speeding IO operation by writing directly to the VMs' memory while bypassing the hypervisor managing the VMs. As such, VFs thwart the dirty page tracking that hypervisors use to minimize VM downtime when the VM is migrated between hosts. The SR-IOV PCIe devices can help resolve this problem by maintaining dirty page tracking data for VMs running on the host machine. The SR-IOV PCIe devices bypassing the hypervisor while writing into a memory page of the VM can set the dirty page tracking data to indicate the memory pages that are dirty (i.e., written to by the VF), and can provide access to the dirty page tracking data. The hypervisor can thereby obtain and use the dirty page tracking data.

Abstract: A method includes providing a neural network having a set of weights. The neural network receives an input data structure for generating a corresponding output array according to values of the set of weights. The neural network is trained to obtain a trained neural network. The training includes setting values of the set of weights with a gradient descent algorithm which exploits a cost function including a loss term and a regularization term. The trained neural network is deployed on a device through a communication network, and used by the device. The regularization term is based on a rate of change of elements of the output array caused by variations of the set of weights values.