Imagination Technologies Limited

James Imber

Adrian Hilton

Cagatay Dikici

Clifford Gibson

Jean-Yves Guillemaut

Szabolcs Cséfalvay

David Walton

Insu Yu

Paul Brasnett

David Hough

Aria Ahmadi

Timothy Smith

Linling Zhang

Chris Martin

Daniel Valdez Balderas

Timothy Atherton

Biswarup Choudhury

Ivaxi Sheth

A method and data processing system implement a neural network containing at least one matrix multiplication operation. The matrix multiplication operation is mapped to a graph of neural network operations including at least one transformation and at least one convolution. The at least one convolution is implemented in fixed-function hardware of a neural network accelerator.

Neural network comprising matrix multiplication

A method of configuring a hardware implementation of a Convolutional Neural Network (CNN), the method comprising: determining, for each of a plurality of layers of the CNN, a first number format for representing weight values in the layer based upon a distribution of weight values for the layer, the first number format comprising a first integer of a first predetermined bit-length and a first exponent value that is fixed for the layer; determining, for each of a plurality of layers of the CNN, a second number format for representing data values in the layer based upon a distribution of expected data values for the layer, the second number format comprising a second integer of a second predetermined bit-length and a second exponent value that is fixed for the layer; and storing the determined number formats for use in configuring the hardware implementation of a CNN.

Convolutional neural network hardware configuration

A data processing system and method are disclosed, for implementing a windowed operation in at least three traversed dimensions. The data processing system maps the windowed operation in at least three traversed dimensions to a plurality of constituent windowed operations in two traversed dimensions. This plurality of 2-D windowed operations is implemented as such in at least one hardware accelerator. The data processing system assembles the results of the constituent 2-D windowed operations to produce the result of the windowed operation in at least three traversed dimensions.

Hardware implementation of windowed operations in three or more dimensions

A method of rendering an image of a 3-D scene includes rendering a noisy image at a first resolution; obtaining one or more guide channels at the first resolution, and obtaining one or more corresponding guide channels at a second resolution. The second resolution may be the same resolution as, or a higher resolution than, the first resolution. For each of a plurality of local neighbourhoods, the method comprises: calculating the parameters of a model that approximates the noisy image as a function of the one or more guide channels (at the first resolution), and applying the calculated parameters to the one or more guide channels at the second resolution, to produce a denoised image at the second resolution.

Rendering an image of a 3-D scene using guided image filtering

Hierarchical methods for selecting fixed point number formats with reduced mantissa bit lengths for representing values input to, and/or output, from, the layers of a DNN. The methods begin with one or more initial fixed point number formats for each layer. The layers are divided into subsets of layers and the mantissa bit lengths of the fixed point number formats are iteratively reduced from the initial fixed point number formats on a per subset basis. If a reduction causes the output error of the DNN to exceed an error threshold, then the reduction is discarded, and no more reductions are made to the layers of the subset. Otherwise a further reduction is made to the fixed point number formats for the layers in that subset. Once no further reductions can be made to any of the subsets the method is repeated for continually increasing numbers of subsets until a predetermined number of layers per subset is achieved.

Hierarchical mantissa bit length selection for hardware implementation of deep neural network

Methods and systems for performing a convolution transpose operation between an input tensor having a plurality of input elements and a filter comprising a plurality of filter weights. The method includes: dividing the filter into a plurality of sub-filters; performing, using hardware logic, a convolution operation between the input tensor and each of the plurality of sub-filters to generate a plurality of sub-output tensors, each sub-output tensor comprising a plurality of output elements; and interleaving, using hardware logic, the output elements of the plurality of sub-output tensors to form a final output tensor for the convolution transpose.

Methods and systems for implementing a convolution transpose layer of a neural network

Hardware implementations of DNNs and related methods with a variable output data format. Specifically, in the hardware implementations and methods described herein the hardware implementation is configured to perform one or more hardware passes to implement a DNN wherein during each hardware pass the hardware implementation receives input data for a particular layer, processes that input data in accordance with the particular layer (and optionally one or more subsequent layers), and outputs the processed data in a desired format based on the layer, or layers, that are processed in the particular hardware pass. In particular, when a hardware implementation receives input data to be processed, the hardware implementation also receives information indicating the desired format for the output data of the hardware pass and the hardware implementation is configured to, prior to outputting the processed data convert the output data to the desired format.

Hardware implementation of a deep neural network with variable output data format

A histogram-based method of selecting a fixed point number format for representing a set of values input to, or output from, a layer of a Deep Neural Network (DNN). The method comprises obtaining a histogram that represents an expected distribution of the set of values of the layer, each bin of the histogram is associated with a frequency value and a representative value in a floating point number format; quantising the representative values according to each of a plurality of potential fixed point number formats; estimating, for each of the plurality of potential fixed point number formats, the total quantisation error based on the frequency values of the histogram and a distance value for each bin that is based on the quantisation of the representative value for that bin; and selecting the fixed point number format associated with the smallest estimated total quantisation error as the optimum fixed point number format for representing the set of values of the layer.

Histogram-based per-layer data format selection for hardware implementation of deep neural network

A method of rendering an image of a 3-D scene includes rendering a noisy image and obtaining one or more guide channels. For each of a plurality of local neighborhoods, the method comprises: calculating the parameters of a model that approximates the noisy image as a function of the one or more guide channels, and applying the calculated parameters to produce a denoised image. At least one of (i) the noisy image, (ii) the one or more guide channels, and (iii) the denoised image, are stored in a quantized low-bitdepth format.

Rendering an image of a 3-D scene

Methods for selecting fixed point number formats for representing values input to and/or output from layers of a DNN which take into account the impact of the fixed point number formats for a particular layer in the context of the DNN. The methods comprise selecting the fixed point number format(s) used to represent sets of values input to and/or output from a layer one layer at a time in a predetermined sequence wherein any layer is preceded in the sequence by the layer(s) from which it depends. The fixed point number format(s) for each layer is/are selected based on the error in the output of the DNN associated with the fixed point number formats. Once the fixed point number format(s) for a layer has/have been selected any calculation of the error in the output of the DNN for a subsequent layer in the sequence is based on that layer being configured to use the selected fixed point number formats.

End-to-end data format selection for hardware implementation of deep neural networks

A method in a hardware implementation of a Convolutional Neural Network (CNN), includes receiving a first subset of data having at least a portion of weight data and at least a portion of input data for a CNN layer and performing, using at least one convolution engine, a convolution of the first subset of data to generate a first partial result; receiving a second subset of data comprising at least a portion of weight data and at least a portion of input data for the CNN layer and performing, using the at least one convolution engine, a convolution of the second subset of data to generate a second partial result; and combining the first partial result and the second partial result to generate at least a portion of convolved data for a layer of the CNN.

Hardware implementation of a convolutional neural network

A method of rendering an image of a 3-D scene includes rendering a noisy image; and obtaining one or more guide channels. For each of a plurality of local neighbourhoods, the method comprises: calculating the parameters of a model that approximates the noisy image as a function of the one or more guide channels, and applying the calculated parameters to produce a denoised image. Tiling is used when calculating the parameters of the model.

Methods for determining a fixed point format for one or more layers of a DNN based on the portion of the output error of the DNN attributed to the fixed point formats of the different layers. Specifically, in the methods described herein the output error of a DNN attributable to the quantisation of the weights or input data values of each layer is determined using a Taylor approximation and the fixed point number format of one or more layers is adjusted based on the attribution. For example, where the fixed point number formats used by a DNN comprises an exponent and a mantissa bit length, the mantissa bit length of the layer allocated the lowest portion of the output error may be reduced, or the mantissa bit length of the layer allocated the highest portion of the output error may be increased. Such a method may be iteratively repeated to determine an optimum set of fixed point number formats for the layers of a DNN.

Error allocation format selection for hardware implementation of deep neural network

Methods and systems for implementing a traditional computer vision algorithm as a neural network. The method includes: receiving a definition of the traditional computer vision algorithm that identifies a sequence of one or more traditional computer vision algorithm operations; mapping each of the one or more traditional computer vision algorithm operations to a set of one or more neural network primitives that is mathematically equivalent to that traditional computer vision algorithm operation; linking the one or more network primitives mapped to each traditional computer vision algorithm operation according to the sequence to form a neural network representing the traditional computer vision algorithm; and configuring hardware logic capable of implementing a neural network to implement the neural network that represents the traditional computer vision algorithm.

Growing community of inventors

Kings Langley, United Kingdom

James Imber