Abstract: Systems and methods for executing software modules in a pipelined fashion. A listing of modules to be executed is received and each module is executed in turn. Prior to execution, each module is code and input checked to determine if it corresponds to a previously executed module. If there is correspondence, then cached results from the previously executed module is used in place of executing the module. If there is no correspondence, then the module is executed, and its results are cached such that these results are available to subsequently executed modules. At least one of the modules may be an implementation of a machine learning model.

Abstract: A method and a system for training a machine learning algorithm (MLA) for object classification. The machine learning algorithm includes an embedding layer and a classification layer. A set of embedding indices representing a reference object is received. The set of embedding indices has been generated based on a byte representation of the reference object. A label associated with the reference object indicative of a reference class the objects belongs to is received. The MLA is iteratively trained to classify objects by embedding the set of embedding indices to obtain an input vector and by predicting an estimated class based on the input vector, and updating a parameter of at least one of the embedding layer and the updated embedding layer. The set of embedding indices is generated by parsing the byte representation to obtain byte n-grams and by applying a hash function on the byte n-grams.

Abstract: A method and a system for generating training images for training an instance segmentation machine learning algorithm (MLA). A set of image-level labelled images are received, where a given image is labelled with a label indicative of a presence of an object having an object class in the image. A classification MLA detects the object having the object class in each image. A class activation map (CAM) indicative of discriminative regions used by the classification MLA for detecting the object in each image is generated. A region proposal MLA is used to generate region proposals for each image. A pseudo mask of the respective object is generated based on the region proposals and the CAM, where a pseudo mask is indicative of pixels corresponding to the respective object class. The pseudo masks are used as a label with the image-level labelled images for training the instance segmentation MLA.

Abstract: There is described a computer-implemented method for generating a vector representation of an image, the computer-implemented method comprising: receiving a given image and semantic information about the given image; generating a first vector representation of the given image using an image embedding method; generating a second vector representation of the semantic information using a word embedding method; combining the first vector representation of the image to be embedded and the second vector representation of the semantic information together, thereby obtaining a modified vector representation for the image to be embedded; and outputting the modified vector representation.

Abstract: There is provided a 2D document extractor for extracting entities from a structured document, the 2D document extractor includes a first convolutional neural network (CNN), a second CNN, and a third recurrent neural network (RNN). A plurality of text sequences and structural elements indicative of location of the text sequences in the document are received. The first CNN encodes the text sequences and structural elements to obtain a 3D encoded image indicative of semantic characteristics of the text sequences and having the structure of the document. The second CNN compresses the 3D encoded image to obtain a feature vector, the feature vector being indicative of a combination of spatial characteristics and semantic characteristics of the 3D encoded image. The third RNN decodes the feature vector to extract the text entities, a given text entity being associated with a text sequence.