Samsung Electronics Co., Ltd.

Other

Mcgill University

Jenkin, Michael

Tarawneh, Enas

Michael Jenkin

Gregory Lewis Dudek

Xue Liu

Di Wu

Xi Chen

Yi Tian Xu

Taeseop Lee

Hang Li

Jikun Kang

Intaik Park

Francois Robert Hogan

David Meger

Enas Tarawneh

John K Tsotsos

Ju Wang

Tianyu Li

Dmitriy Rivkin

Yogesh Girdhar

Chris Prahacs

Sahand Rezaei-Shoshtari

Shane Saunderson

Philippe Giguere

Jimmy Li

Junaed Sattar

Kaleem Siddiqi

Charlotte Morissette

Affan Jilani

Greg Dudek

A server may receive a first traffic data and a second traffic data from a first base station and a second base station; obtain a first augmented traffic data for the first base station, based on the first traffic data and a subset data of the second traffic data; obtain a second augmented traffic data for the second base station, based on the second traffic data and a subset data of the first traffic data; obtain a first artificial intelligence (AI) model via imitation learning based on the first augmented traffic data; obtain a second AI model imitation learning based on the second augmented traffic data; obtain a generalized AI model via knowledge distillation from the first AI model and the second AI model; and predict a future traffic load of each of the first base station and the second base station based on the generalized AI model.

Multi-batch reinforcement learning via multi-imitation learning

A method performed by an optical tactile sensor comprising a camera, includes: measuring a plurality of relations between the optical tactile sensor and an object; detecting a contact of a membrane on the object; detecting a plurality of markers on the membrane; performing a plurality of first operations by using a tactile perception module; and performing a second operation by using a three-dimensional (3D) perception module, based on a result of the performing of the plurality of first operations.

Optical tactile sensor and method for estimating shape from touch

A method of forecasting a future load may include: obtaining source data sets and a target data set that have been collected from a plurality of source base stations and a target base station, respectively; among a plurality of source machine learning models, selecting at least one machine learn source model that has a traffic load prediction performance higher than that of a target machine learning model through a negative transfer analysis; obtaining model weights to be applied to the target machine learning model and the selected at least one source machine learning model via an attention neural network that is jointly trained with the target machine learning model and the selected source machine learning models; obtaining a load forecasting model for the target base station by combining the target machine learning model and the selected at least one source machine learning model according to the model weights; and predicting a future communication traffic load of the target base station based on the load forecasting model.

Method of load forecasting via attentive knowledge transfer, and an apparatus for the same

Rapid and data-efficient training of an artificial intelligence (AI) algorithm are disclosed. Ground truth data are not available and a policy must be learned based on limited interactions with a system. A policy bank is used to explore different policies on a target system with shallow probing. A target policy is chosen by comparing a good policy from the shallow probing with a base target policy which has evolved over other learning experiences. The target policy then interacts with the target system and a replay buffer is built up. The base target policy is then updated using gradients found with respect to the transition experience stored in the replay buffer. The base target policy is quickly learned and is robust for application to new, unseen, systems.

Load balancing using data-efficient learning

A method, computer program, and computer system are provided for load forecasting. Datasets corresponding to source machine learning models and a target domain base model are identified. A set of forecasting models corresponding to the identified datasets are learned. An ensemble model is determined from the learned set of forecasting models based on gradient boosting. An available resource is allocated based on the ensemble model.

Short-term load forecasting

An apparatus distributing communication load over a plurality of communication cells may select action centers from random cell reselection values, based on a standard deviation of an internet protocol (IP) throughout over the plurality of communication cells; input a first vector indicating a communication state of a communication system and a second vector indicating the standard deviation of the IP throughout of the plurality of communication cells, to a neural network to output a sum of the action centers and offsets as cell reselection parameters; and transmit the cell reselection parameters to the communication system to enable a base station of the communication system to perform a cell reselection based on the cell reselection parameters.

Method and apparatus for distributing traffic load between different communication cells

A method of sensing an interaction with an object using a semitransparent tactile surface (STS) sensor having an image sensor and a semitransparent membrane includes capturing, by the image sensor, a stream of images; separating the stream of images into a visual stream including light traveling through the semitransparent membrane and a tactile stream including light reflecting off of the semitransparent membrane; and processing the visual stream and tactile stream through a multimodal deep neural network.

Semitransparent tactile surface sensor and a method of sensing an interaction with an object using the semitransparent tactile surface sensor

There are provided systems and methods for rendering of an animated avatar. An embodiment of the method includes: determining a first rendering time of a first clip as approximately equivalent to a predetermined acceptable rendering latency, a first playing time of the first clip determined as approximately the first rendering time multiplied by a multiplicative factor; rendering the first clip; determining a subsequent rendering time for each of one or more subsequent clips, each subsequent rendering time is determined to be approximately equivalent to the predetermined acceptable rendering latency plus the total playing time of the preceding clips, each subsequent playing time is determined to be approximately the rendering time of the respective subsequent clip multiplied by the multiplicative factor; and rendering the one or more subsequent clips.

System and method for rendering of an animated avatar

A control system for a robotic device maneuverable in at least a liquid medium, the system having at least one visual sensor retrieving an image of the device's environment, an image analyzing module receiving the image, determining a presence of an object of a given type therein and analyzing at least one property of the object, a motion calculator determining a desired motion of the device based on the property, and a controller operating a propulsion system of the device to obtain the desired motion. Also, a legged robotic device having a control system including at least one sensor providing data about an environment of the device, the control system using sensor data to determine a desired motion of the device, determining a corresponding required leg motion of each of the legs to produce the desired motion and actuating the legs in accordance with the corresponding required leg motion.

Amphibious robotic device

A video surface is constructed by adjoining a large number of flat screen display devices together. Each screen on this surface is controlled by its own computer processor and these processors are networked together. Superimposed over this surface is a tiling of transparent touch-sensitive screens which allow for user input. The resulting display is thin, has a very high resolution, appears to be a single large screen to the user, and is capable of supporting many different types of human-machine interaction.

Large-scale, touch-sensitive video display

Growing community of inventors

Toronto, Canada

Michael Jenkin