Abstract: A method and a system for recommending items in an online listing platform are provided. The method comprises: receiving, from an electronic device, a request for item recommendations triggered by the user having interacted with a given item of the online listing platform; identifying, in the plurality of items, based on the given item, a set of recommendable items; obtaining user data of the user; determining, based on the user data, for a given recommendable item, a respective value of a user-specific ranking parameter, which is indicative of a likelihood value of the user perceiving the given recommendable item as being of interest to the user; ranking the set of recommendable items according to respective values of the user-specific ranking parameter, thereby generating a ranked set of recommendable items; and selecting, from the ranked set of recommendable items, at least one recommendable for presentation thereof to the user.

Abstract: A method and a system for recommending items in an online listing platform are provided. The method comprises: receiving, from an electronic device, a request for item recommendations triggered by the user having interacted with a given item of the online listing platform; identifying, in the plurality of items, based on the given item, a set of recommendable items; obtaining user data of the user; determining, based on the user data, for a given recommendable item, a respective value of a user-specific ranking parameter, which is indicative of a likelihood value of the user perceiving the given recommendable item as being of interest to the user; ranking the set of recommendable items according to respective values of the user-specific ranking parameter, thereby generating a ranked set of recommendable items; and selecting, from the ranked set of recommendable items, at least one recommendable for presentation thereof to the user.

Abstract: There is disclosed a method and system for classifying a word as an obscene word, the method comprising, at a training phrase: acquiring a first word, the first word corresponding to a given obscene word; generating a first set of misspelled words, the first set of misspelled words comprising a plurality of misspelled variations of the first word; generating a training pairs, the training pairs comprising: a set of positive training pairs comprising the first word paired with each misspelled variations of the first word; training a machine learning algorithm, the training comprising: determining, for each training pairs, a set of features representative of a property of the training pairs; generating an inferred function based on the set of features, the inferred function being configured to assign, in use, an indecency score, the decency score being indicative of a likelihood of the word being obscene.

Abstract: A method and system for determining rank positions of elements displayed on a search engine results page (SERP) by a ranking system. During a training phase of the ranking system, an indication of user interactions associated with a first ranked list of elements is acquired. The elements of the first ranked list were visually displayed on a first SERP. A score for a first content element is determined based on a size of the first content element and a ranking of the first content element. The ranking system is trained, based on the score, to predict a predicted usefulness score for the first content element. During an in-use phase of the ranking system, a ranked list of elements is determined based on a query. The ranking system determines a ranked position for a content element in the ranked list. A SERP based on the ranked list is output.

Abstract: Methods and devices for generating data for controlling a Self-Driving Car (SDC) are disclosed. The method includes: i) receiving a section of a road map corresponding to surroundings of the SDC and at least one object, ii) generating a plurality of predicted trajectories including a potential future location points of the at least one object, iii) mapping the potential future location points on the section of the road map, iv) computing a score for each of the potential future location points, the score representing an association of a given potential future location point with the plurality of road lanes at a future instance of time, v) computing an aggregated score from the scores corresponding to potential future location points, and vi) based on the aggregated score, determining a predicted location of the at least one object at the future instance of time.

Abstract: An optical system and method for controlling the system, and a self-driving vehicle connected thereto. The optical system includes a housing; a first light source configured to emit light in a first wavelength band; a second light source configured to emit light in a second wavelength band different from the first wavelength band; a dichroic element configured for combining light from the first light source and the second light source into an output light beam to be scanned over a field of view; a scanning unit configured to direct output beam outwardly from the optical system, the scanning unit being disposed in the housing; a controller communicatively coupled with the scanning unit; and at least one sensor communicatively coupled with the controller and configured to sense light reflected off surrounding objects into the optical system.

Abstract: Methods and devices for operating a Self-Driving Car (SDC) are disclosed. The method includes generating a first graph-structure having nodes and edges, ranking the edges based on a priority logic into a ranked list of edges, and generating a second graph-structure (i) by iteratively generating attributes for respective ones from the ranked list of edges beginning with a highest priority edge in the ranked list of edges and (ii) until a pre-determined limit is met. The method also includes causing operation of the SDC on the road segment using the second graph-structure.

Abstract: An assembly for a wheel of a robotic vehicle and a method for overcoming an obstacle for said robotic vehicle. The assembly comprises a first arm portion and a second arm portion, the first arm portion being attachable to a chassis of the robotic vehicle at an attachment point and extending forwardly and downwardly relative to the attachment point in a direction of a movement of the robotic vehicle. The second arm portion is pivotably connected with the first arm portion at an arm pivot point and extends forwardly relative to the arm pivot point in the direction of the movement. The wheel is rotatably mounted on one end of the second arm portion opposed to the arm pivot point. A wheel rotation axis being positioned at least as high as the arm pivot point relative to a surface on which the robotic vehicle is positioned.

Abstract: There is disclosed a method of and a system for training and using a Machine Learning Algorithm (MLA), the MLA using a decision tree model having a decision tree. During training a training object being associated with a categorical feature and is processed at a node of the decision tree. The method comprises calculating a numeric representation of the categorical feature and the value of the splits for the node “in-line” with generating a given iteration of the decision tree.

Abstract: A method and system for providing scan data for decision making in a self-driving vehicle are provided. The self-driving vehicle includes a scanning sensor that scans a space around the self-driving vehicle in consecutive scanning cycles. The method includes receiving a request for scan information associated with a requested segment of space. The method determines whether to wait for a next scan of the requested segment and use scan data from the next scan, or to extract and use scan data from a previous scan of the requested segment. The method also includes processing the scan data to generate the scan information for the requested segment, providing the scan information for the requested segment in response to the request, and operating the self-driving vehicle in accordance with a decision made about a maneuver to be performed by the self-driving vehicle based, at least in part, on the scan information.

Abstract: Methods and electronic devices for executing offline translation of a source word into a target word via a Neural Network an encoder and a decoder. The method includes splitting the source word into input tokens, generating vector representations for input tokens, and generating a first sequence of output tokens representative of a first candidate word. In response to the first candidate word not respecting at least one pre-determined rule, the method includes triggering the decoder to generate a second sequence of output tokens having a different at least one last output token than at least one last output token of the first sequence. The second sequence is representative of a second candidate word. In response to the second candidate word respecting the at least one pre-determined rule, the method includes determining that the second candidate word is the target word.

Abstract: Method and server for training an Attention-based Neural Network (ANN) to generate a textual output sequence being a content summary to be used as a response to a query. The ANN has an encoder and a decoder sub-network. The method includes, (i) inputting a query and input sequence into the encoder sub-network where the input sequence is a sequence of input groups associated with corresponding content snippets, (ii) generating an encoded representation of the input sequence, which includes generating attention-type outputs for respective words from the input sequence by applying an attention-limiting mask configured to attend only to words from the given input group, (iii) generating a decoded representation being a predicted textual output sequence, (iv) generating a penalty and (iii) adjusting the ANN based on the penalty score.

Abstract: Methods and electronic devices for processing a spoken utterance associated with a user of an electronic device are disclosed. The method includes generating a textual representation of the spoken utterance having words, identifying a nonce word and a non-normalized word amongst the words, and generating a plurality of candidate textual representations based on the textual representation. The candidates have at least one of a first set of candidate textual representations and a second set of candidate textual representations, such that candidates from the first set are missing the nonce word from the words of the textual representation, and candidates from the second set have the non-normalized word from the words of the textual representation replaced by a normalized version thereof. The method includes comparing the candidates against grammars, and in response to a match, triggering an action associated with the grammar.

Abstract: A method and a system for augmenting an audio feed for inclusion of data therein suitable for determining an identity of a human assessor are provided. The method comprises: receiving the audio feed; receiving an indication of identity of the human assessor to whom the audio feed is to be transmitted, the indication of identity being representable by a unique sequence of bits; generating, based on the unique sequence of bits, an identity watermark associated with the human assessor to be included in the audio feed to generate an augmented audio feed, by modifying the audio signal to have predetermined energy level at each of at least two different frequency levels to indicate presence of the given bit of the unique sequence of bits associated with the human assessor in the augmented audio feed; and transmit the augmented audio feed to an electronic device associated with the human assessor.

Abstract: A method and a system for training a machine-learning algorithm (MLA) to determine a user class of a user of an electronic device are provided. The method comprises: receiving a training audio signal representative of a training user utterance; soliciting, by the processor, a plurality of assessor-generated labels for the training audio signal, the given one of the plurality of assessor-generated labels being indicative of whether the training user is perceived to be one of a first class and a second class; generating an amalgamated assessor-generated label for the training audio signal, the amalgamated assessor-generated label being indicative of a label distribution of the plurality of assessor-generated labels between the first class and the second class; generating a training set of data including the training audio signal and the amalgamated assessor-generated to train the MLA to determine the user class of the user producing an in-use user utterance.

Abstract: Methods and servers for storing data associated with users and digital items of a recommendation system having access to non-distributed and distributed storages. The server trains a model based for generating first user and item embeddings. The server stores (i) the first user embeddings in the non-distributed storage, and (ii) the first item embeddings in the distributed storage. The server re-trains the model for generating second user and item embeddings. The server stores (i) the second user embeddings in the non-distributed storage in addition to the first user embeddings, and (ii) second item embeddings in the distributed storage instead of the respective first item embeddings by replacing the respective first item embeddings. When the second item embeddings are stored on each node of the distributed storage, the server removes the first user embeddings associated with the first value from the non-distributed storage.