Intel Corporation

Takadu Ltd.

Amitai Armon

Lev Faivishevsky

Tomer Schwartz

Yaniv Fais

Jacob Subag

Uzi Sarel

Jeremie Dreyfuss

Tomer Bar-On

Ehud Cohen

Yahav Shadmiy

Amit Bleiweiss

Gal Leibovich

Haggai Scolnicov

Michael Behar

Itamar Ben-Ari

Chaim Linhart

Guy Jacob

Gal Novik

Lilach Bien

Raanan Yonatan Yehezkel Rohekar

Shai Gutner

Amir Peleg

Uri Barkay

Raz Ziv

Justin E Gottschlich

Niranjan Hasabnis

Ashwin Muppalla

Tahi Hollander

Sanket Save

Oren David Kimhi

Subash Sudireddy

Moty Fania

Noam Petrank

In an example, an apparatus comprises a plurality of execution units comprising and logic, at least partially including hardware logic, to receive a plurality of data inputs for training a neural network, wherein the data inputs comprise training data and weights inputs; represent the data inputs in a first form; and represent the weight inputs in a second form. Other embodiments are also disclosed and claimed.

Real time context dependent deep learning

In an example, an apparatus comprises logic, at least partially including hardware logic, to implement a lossy compression algorithm which utilizes a data transform and quantization process to compress data in a convolutional neural network (CNN) layer. Other embodiments are also disclosed and claimed.

Low rank matrix compression

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 expose embedded cast operations in at least one of a load instruction or a store instruction; determine a target precision level for the cast operations; and load the cast operations at the target precision level. Other embodiments are also disclosed and claimed.

Variable precision and mix type representation of multiple layers in a network

A mechanism is described for facilitating memory handling and data management in machine learning at autonomous machines. A method of embodiments, as described herein, includes detecting multiple tables associated with multiple neural networks at multiple autonomous machines, where each of the multiple tables include an index. The method may further include combining the multiple tables and multiple indexes associated with the multiple tables into a single table and a single index, respectively, where the single table is communicated to the multiple autonomous machines to allow simultaneous processing of one or more portions of the single table using one or more memory devices and one or more processors of one or more of the multiple autonomous machines.

Smart memory handling and data management for machine learning networks

Methods, apparatus, systems, and articles of manufacture are disclosed to detect code defects. An example apparatus includes repository interface circuitry to retrieve code repositories corresponding to a programming language of interest, tree generating circuitry to generate parse trees corresponding to code blocks contained in the code repositories, directed acyclic graph (DAG) circuitry to generate DAGs corresponding to respective ones of the parse trees, the DAGs including control flow information and data flow information, abstraction generating circuitry to abstract the DAGs, invariant identification circuitry to extract invariants from the abstracted DAGs, and DAG comparison circuitry to cluster respective ones of the extracted invariants to identify respective ones of the abstracted DAGs with common invariants.

Methods, systems, articles of manufacture and apparatus to detect code defects

In an example, an apparatus comprises logic, at least partially including hardware logic, to implement a lossy compression algorithm which utilizes a data transform and quantization process to compress data in a convolutional neural network (CNN) layer.

In an example, an apparatus comprises a plurality of execution units; and logic, at least partially including hardware logic, to determine a sub-graph of a network that can be executed in a frequency domain and apply computations in the sub-graph in the frequency domain. Other embodiments are also disclosed and claimed.

Sub-graph in frequency domain and dynamic selection of convolution implementation on a GPU

A mechanism is described for facilitating visual anomaly detection without reference in computing environments. An apparatus of embodiments, as described herein, includes one or more processors to select a frame from a sequence of multiple frames associated with a video stream captured by a camera, and dynamically compute a frame confidence score for the frame based on frame training data associated with frame. The one or more processors are further to detect one or more anomalies in the frame when the frame confidence score is less than a frame confidence threshold associated with the frame, where detecting includes dynamically comparing the frame confidence score with the frame confidence threshold through inference using frame field data and the frame training data.

Visual anomaly detection without reference in graphics computing environments

Examples of systems and methods for multisensory change detection are generally described herein. A method may include receiving a first set of signals from a first combination of sensors and a second set of signals from a second combination of sensors in a plurality of sensors, and determining a first distribution for the first set of signals and a second distribution for the second set of signals. The method may include estimating a divergence between the first and second distributions using the first and second combinations of sensors, a count of the plurality of sensors, and distances from a plurality of signals in the second set of signals to a first plurality of nearest neighbor signals in the first set of signals and a second plurality of nearest neighbor signals in the second set of signals. The method may include determining whether the divergence exceeds a threshold.

Multisensory change detection for internet of things domain

Technologies for optimization of machine learning training include a computing device to train a machine learning network with a training algorithm that is configured with configuration parameters. The computing device may perform many training instances in parallel. The computing device captures a time series of partial accuracy values from the training. Each partial accuracy value is indicative of machine learning network accuracy at an associated training iteration. The computing device inputs the configuration parameters to a feed-forward neural network to generate a representation and inputs the representation to a recurrent neural network. The computing device trains the feed-forward neural network and the recurrent neural network against the partial accuracy values. The computing device optimizes the feed-forward neural network and the recurrent neural network to determine optimized configuration parameters. The optimized configuration parameters may minimize training time to achieve a predetermined accuracy level. Other embodiments are described and claimed.

Technologies for optimized machine learning training

Growing community of inventors

Tel Aviv, Israel

Amitai Armon