Abstract: The prevention of vehicle accidents is targeted. A road texture model is created based on a vehicle camera image. An initial vehicle location estimate is determined, and map imagery is obtained based on this location estimate. A refined vehicle location is determined using visual egomotion. In particular, 3D features of the vehicle image and the retrieved map imagery are identified and aligned. A map image is selected based on this alignment, and the location associated with the map image is modified by a displacement between the selected map image and the vehicle image to produce a refined vehicle location. A road boundary model is created based on the road texture model and the refined vehicle location, and a road departure model is created based on the road boundary model and vehicle odometry information. The operator of the vehicle is warned of a road departure based on the road departure model.

Abstract: The present invention provides a method for incorporating features from heterogeneous auxiliary datasets into input text data for use in classification. Heterogeneous auxiliary datasets, such as labeled datasets and unlabeled datasets, are accessed after receiving input text data. Features are extracted from each of the heterogeneous auxiliary datasets. The features are combined with the input text data to generate a set of features which may potentially be used to classify the input text data. Classification features are then extracted from the set of features and used to classify the input text data. In one embodiment, the classification features are extracted by calculating a mutual information value associated with each feature in the set of features and identifying features having a mutual information value exceeding a threshold value.

Abstract: The invention relates to selecting a set of candidate genes from a pool of genes. The method comprising receiving a set of gene data; arranging the set of gene data into a set of clusters with similar profiles by use of a clustering algorithm; and inputting the set of clusters into a genetic algorithm to select a set of candidate genes from the set of clusters. The method thus relates to hybrid between selection by clustering computation and selection by evolutionary computation. This hybrid is also referred to as an evolutionary clustering algorithm (ECA).

Abstract: A room temperature crosslinkable polymer system comprising an anhydride containing polymer and an oxyalkylene amine and a polymer electrolyte derived therefrom are prepared and employed as ion conducting materials for batteries such as lithium ion battery, solar cells and electrochromic devices is disclosed.

Abstract: The present invention provides a method for tracking entities, such as people, in an environment over long time periods. A region-based model is generated to model beliefs about entity locations. Each region corresponds to a discrete area representing a location where an entity is likely to be found. Each region includes one or more positions which more precisely specify the location of an entity within the region so that the region defines a probability distribution of the entity residing at different positions within the region. A region-based particle filtering method is applied to entities within the regions so that the probability distribution of each region is updated to indicate the likelihood of the entity residing in a particular region as the entity moves.

Abstract: A system and method enable an autonomous machine such as an indoor humanoid robot to systematically process user commands and respond to situations. The method captures distributed knowledge from human volunteers, referred to as “commonsense knowledge.” The commonsense knowledge comprises classes such as steps for tasks, responses to situations, and locations and uses of objects. Filtering refines the commonsense knowledge into useful class rules. A second level of rules referred to as meta-rules performs reasoning by responding to user commands or observed situations, orchestrating the class rules and generating a sequence of task steps. A task sequencer processes the generated task steps and drives the mechanical systems of the autonomous machine.

Abstract: A method for controlling receive diversity of an antenna system of a computer device, the antenna system including two or more antenna elements. The method includes establishing a session with a remote transmitting system and determining whether a predetermined criteria detected by the computer device is satisfied within the session. If the predetermined criteria is satisfied within the session, the method includes activating at least two of the antenna elements for receiving transmissions, enabling performance of receive diversity on the received transmissions, and performing receive diversity on the received transmissions. If the predetermined criteria is not satisfied within the session, the method includes activating at least one of the antenna elements, disabling performance of receive diversity on the received transmissions, and performing a default signal processing on the received transmissions. A mobile communication device may be used to perform the method.

Abstract: Extraction of data employing a sequence of remote function calls in form of a synchronous multi-batch call chain is provided. Sequencing of calls is enabled by generating parameters associated with the extraction of a next batch of data. The parameterized, dynamic generation of queries allows for optimization of memory utilization by batching result sets and data conversion. Each subsequent call retrieves a packet of data, picking up where the previous call left off without an overlap. Parameters are updated after each call based on extracted data, available memory, and the like.

Abstract: A method and system for storing episodic sequences (events and actions). The system learns episodic sequencing by observing real-world events and actions or by receiving fact data from a database storing common sense facts. The episodic sequences are classified into events and actions, processed to indicate correlations and causality between the events and actions, and generated into linked graphs. The linked graphs may then be used to draw inferences, recognize patterns, and make decisions.

Abstract: The invention includes a computer based system or method for automatically generating a grammar associated with a first task comprising the steps of: receiving first data representing the first task based from responses received from a distributed network; automatically tagging the first data into parts of speech to form first tagged data; identifying filler words and core words from said first tagged data; modeling sentence structure based upon said first tagged data using a first set of rules; identifying synonyms of said core words; and creating the grammar for the first task using said modeled sentence structure, first tagged data and said synonyms.

Abstract: Disclosed is a flame-retardant composition comprising (i) 40-66 wt. % alkenyl aromatic resin, (ii) 9-33 wt. % ammonium polyphosphate and (iii) 14-40 wt. % cellulosic material, wherein all weights are based on the total weight of the composition and wherein ammonium polyphosphate and cellulosic material are present in a weight % ratio effective to provide molded articles exhibiting at least V-1 flame rating as determined according to the UL-94 protocol. A process to prepare the composition and articles comprising a composition of the invention and/or made by the process of the invention described herein are also disclosed.

Abstract: Simultaneous localization and mapping (SLAM) utilizes multiple view feature descriptors to robustly determine location despite appearance changes that would stifle conventional systems. A SLAM algorithm generates a feature descriptor for a scene from different perspectives using kernel principal component analysis (KPCA). When the SLAM module subsequently receives a recognition image after a wide baseline change, it can refer to correspondences from the feature descriptor to continue map building and/or determine location. Appearance variations can result from, for example, a change in illumination, partial occlusion, a change in scale, a change in orientation, change in distance, warping, and the like. After an appearance variation, a structure-from-motion module uses feature descriptors to reorient itself and continue map building using an extended Kalman Filter.

Abstract: The present invention provides a method for identifying a turn, such as a sentence or phrase, for addition to a platform dialog comprising a plurality of turns. Lexical features of each of a set of candidate turns relative to one or more turns in the platform dialog are determined. Semantic features associated with each candidate turn and associated with the platform dialog are determined to identify one or more topics associated with each candidate turn and with the platform dialog. Lexical features of each candidate turn are compared to lexical features of the platform dialog and semantic features associated with each candidate turn are compared to semantic features of the platform dialog to rank the candidate turns based on similarity of lexical features and semantic features of each candidate turn to lexical features and semantic features of the platform dialog.

Abstract: A system, apparatus and application for providing robots with the ability to intelligently respond to perceived situations are described. A knowledge database is assembled automatically, based on distributed knowledge capture. The knowledge base embodies the “common sense,” that is, the consensus, of the subjects who contribute the knowledge. Systems are provided to automatically preprocess, or “clean” the information to make it more useful. The knowledge thus refined is utilized to construct a multidimensional semantic network, or MSN. The MSN provides a compact and efficient semantic representation suitable for extraction of knowledge for inference purposes and serves as the basis for task and response selection. When the robot perceives a situation that warrants a response, an appropriate subset of the MSN is extracted into a Bayes network. The resultant network is refined, and used to derive a set of response probabilities, which the robot uses to formulate a response.

Abstract: Recurring appointment items in a calendaring application are synchronized and recorded in a backend Line Of Business (LOB) service. Items marked in the calendaring application for time recording are used to create a recurring bound item with extended properties. A sweeper module scans on every run the mailbox for such recurring bound items. Occurrences of bound recurring appointments, which occur between the last scan and the current scan, are determined and detached from the corresponding recurring series, such that standalone bound items, are created for such occurrences that can be recognized by an LOB application. Entries for bound items are retrieved from a client data store and synchronized it with the backend LOB service through a create, update, or delete operation request.

Abstract: Secure sharing of bound information is enabled in client applications associated with a backend LOB service. Bound item IDs are assigned to newly created bound items by a client, the items synchronized with the LOB system, a correlation ID received in response to the synchronization, and the two IDs mapped. A reverse sequence of actions is performed when the LOB service creates the bound item. In response to an attempt by a client to exchange bound information, the item is placed in a pending state and allowed to be received by a receiver upon successful completion of data transfer and LOB system permission of the exchange passing the bound item ID and the correlation ID. If the LOB system rejects the exchange, the item is placed in an unbound state and the receiver not allowed to receive the bound information.

Abstract: The present invention provides a method for learning a policy used by a computing system to perform a task, such delivery of one or more objects by the computing system. During a first time interval, the computing system determines a first state, a first action and a first reward value. As the computing system determines different states, actions and reward values during subsequent time intervals, a state description identifying the current sate, the current action, the current reward and a predicted action is stored. Responsive to a variance of a stored state description falling below a threshold value, the stored state description is used to modify one or more weights in the policy associated with the first state.

Abstract: A method and system for obtaining reverse channel resources for a mobile device, the method filtering a closed loop correction power value; calculating a data rate utilizing the filtered closed loop correction power value; and sending a supplemental channel request message, including the calculated data rate, to a base station transceiver system.

Abstract: Architecture that maintains the user experience as close as possible to the user experience when dealing with the usual/regular forms and data provided in an office application when dealing with hybrid forms and hybrid data. Synchronization of the hybrid data to single-master and multi-masters systems is accomplished when the associated hybrid client goes offline, changes are made to the hybrid data, and then the client comes back online. For example, where the single-master system is a line-of-business (LOB) server system and the multi-master system is an officer server that employs collaboration capability, the architecture allows office users to access, manipulate, and share LOB entity information using collaborative means, while at the same time, ensuring data and business process consistency for the LOB entity in the LOB system.

Abstract: A system and a method are disclosed for automatic question classification and answering. A multipart artificial neural network (ANN) comprising a main ANN and an auxiliary ANN classifies a received question according to one of a plurality of defined categories. Unlabeled data is received from a source, such as a plurality of human volunteers. The unlabeled data comprises additional questions that might be asked of an autonomous machine such as a humanoid robot, and is used to train the auxiliary ANN in an unsupervised mode. The unsupervised training can comprise multiple auxiliary tasks that generate labeled data from the unlabeled data, thereby learning an underlying structure. Once the auxiliary ANN has trained, the weights are frozen and transferred to the main ANN. The main ANN can then be trained using labeled questions. The original question to be answered is applied to the trained main ANN, which assigns one of the defined categories.