首页期刊导航|Robotics & Machine Learning Daily News
期刊信息/Journal information
Robotics & Machine Learning Daily News
NewsRx
Robotics & Machine Learning Daily News

NewsRx

Robotics & Machine Learning Daily News/Journal Robotics & Machine Learning Daily News
正式出版
收录年代

    Patent Issued for Bio-behavior system and method (USPTO 12035136)

    131-134页
    查看更多>>摘要:Patent number 12035136 is assigned to SecureAuth Corporation (Irvine, California , United States).The following quote was obtained by the news editors from the background informa tion suppliedby the inventors: “Digital transactions of a variety of types may stem not from a party authorized toenter into the transaction but by parties th at are either unauthorized to enter into the transaction orbad actors and netwo rk bots who have acquired the means to enter into the transaction illegally froma hostile environment. The hostile environment that may have resulted from a De nial of Service (DoS)attack from sources such as User Datagram Protocol (UDP) f looding, Internet Control Message Protocol(ICMP) flooding, and/or Portscan. For instance, a stolen credit card number or bank account access maybe utilized to make fraudulent purchases or transactions-exchanges. A stolen or compromised pa sswordmay be utilized to improperly access information. Even conventional purch ases or activities within anorganization may be engaged in by an employee or me mber who does not have authorization to do so.”

    'Laser Dicing System For Filamenting And Singulating Optical Devices' in Patent Application Approval Process (USPTO 20240227081)

    134-137页
    查看更多>>摘要:This patent application has not been assigned to a company or institution.The following quote was obtained by the news editors from the background informa tion supplied bythe inventors: ““Field“Embodiments of the present disclosure generally relate to a laser dicing system for optical devices.“Description of the Related Art.

    Researchers Submit Patent Application, 'Systems And Methods For Dataset Vector S earc

    137-141页
    查看更多>>摘要:The patent’s assignee is Snark AI Inc. (San Francisco, California, United States ).News editors obtained the following quote from the background information suppli ed by the inventors:“Machine-learning datasets can be both large and varied, in cluding large amounts of information in severaldifferent formats. The large siz e, complexity, and format of the dataset can create technical difficulties inma naging and communicating the dataset between systems, storing the dataset, proce ssing the dataset,and utilizing the dataset in machine-learning processes.”

    Patent Application Titled 'System And Method For Operational Analysis Of Energy Storage Devices' Published Online (USPTO 20240232694)

    141-145页
    查看更多>>摘要:No assignee for this patent application has been made.Reporters obtained the following quote from the background information supplied by the inventors:“Energy storage devices are commonly used for electronic devic es throughout society, including for electricvehicles, powering homes, etc. Ene rgy storage devices, such as a battery, may be reused after their firstlife to reduce the environmental impact of disposal and to increase the economic benefit for each battery.”

    'Convertible Ride-On and Walk-About Platform for A Robotic Upper Exoskeleton' in Patent Application Approval Process (USPTO 20240225942)

    146-150页
    查看更多>>摘要:This patent application has not been assigned to a company or institution.The following quote was obtained by the news editors from the background informa tion supplied by theinventors: “A wide variety of exoskeleton, humanoid, roboti c arms, and other robots or robotic systemsexist which perform tasks in a varie ty of situations and applications. Robotic exoskeletons in particular arewearab le electromechanical devices that have been developed as augmentative devices to assist, enhance,or amplify the physical performance of the wearer or as orthot ic devices for gait rehabilitation or locomotionassistance. Robotic exoskeleton s have potential applications in multiple different fields and may be usedby a variety of different operators. While many exoskeleton systems comprise an upper body exoskeletonportion supported by a lower body exoskeleton portion (e.g., o ne comprising two legs) that interfaces withthe lower body of a human operator upon the operator donning the exoskeleton, lower body exoskeletonsare often com plex in their configuration by comprising multiple actuatable joints to facilita te movement inmultiple degrees of freedom that resemble as closely as possible the kinematics of the human operator dueto these being physically coupled to th e human operator, namely to the legs of the human operator. Thislevel of comple xity within a lower body exoskeleton portion in many instances is not necessary. Indeed,there are many instances where the types of amplified or assisted movem ents and/or maneuvers that ahuman operator may need to perform with the lower b ody exoskeleton portion in order to complete one ormore tasks with the upper bo dy exoskeleton are simple, thus rendering a complex lower body exoskeletonporti on unnecessary in that it possesses a much greater capability than what is neede d. Another wayof stating this is that a complex exoskeleton may possess actuata ble joints, degrees of freedom andvarious components, elements and systems need ed to operate the lower body exoskeleton portion as it isinterfaced with a huma n operator that are only there due to the complex configuration of the lower body exoskeleton to enable the human interface. However, in reality, such complexit ies may be overkill formany tasks that need to be carried out using the upper b ody exoskeleton portion. While a complex lowerbody exoskeleton portion can cert ainly perform simple amplified or movements and/or maneuvers, suchas serving as a support for the upper exoskeleton portion as interfaced with an upper body of the user,facilitating amplified lifting, squatting, bending over, walking from one location to another, etc. by theoperator, it likely does these utilizing a ll available systems, components, degrees of freedom, etc. withinthe lower body exoskeleton portion. In addition, it is recognized that in some instances a low er bodyexoskeleton portion may merely serve as a support for the upper exoskele ton portion, such as for a taskthat can be undertaken and accomplished using on ly the upper body exoskeleton portion. A complex lowerbody exoskeleton can be c ostly and can also be cumbersome, particularly when performing lower bodymoveme nts and/or maneuvers, such as walking, squatting, and even standing for an exten d

    Patent Issued for Adaptive selection of machine learning/deep learning model wit h optimal hyper-parameters for anomaly detection of connected equipment (USPTO 1 2032344)

    150-153页
    查看更多>>摘要:The patent’s assignee for patent number 12032344 is Tyco Fire & Se curity GmbH (Neuhausen amRheinfall, Switzerland).News editors obtained the following quote from the background information suppli ed by the inventors:“The present disclosure relates generally to chillers opera ting in conjunction with HVAC systems. Thepresent disclosure relates more parti cularly to predicting chiller faults using models trained with machinelearning and deep learning.

    Patent Application Titled 'Systems And Methods For Automatic Data Annotation' Pu blished Online (USPTO 20240233419)

    153-157页
    查看更多>>摘要:The assignee for this patent application is Shanghai United Imaging Intelligence Co. Ltd. (Shanghai,People’s Republic of China).Reporters obtained the following quote from the background information supplied by the inventors:“Having annotated data is crucial to the training of machine-l earning (ML) models or artificial neuralnetworks. Conventional ways of data ann otation rely heavily on manual work (e.g., by qualified annotatorssuch as radio logists if the data includes medical images) and even when computer-based tools are provided,they still require a tremendous amount of human effort (e.g., mous e clicking, drag-and-drop, etc.). Thisstrains resources and often leads to inad equate and/or inaccurate results. Accordingly, it is highly desirableto develop systems and methods to automate the data annotation process such that more data may beobtained for ML training and/or verification.”

    Researchers Submit Patent Application, 'Systems and Methods for Producing a Prod uct', for Approval (USPTO 20240232805)

    157-162页
    查看更多>>摘要:News editors obtained the following quote from the background information suppli ed by the inventors:“Many attributes relate to a product. For example, products are formed of compositions, such as chemicalcompositions. Chemical composition s typically include many different ingredients. Each of the ingredientshas a pa rticular value, such as a chemoinformatic value, associated with the ingredient. Further, oneor more properties (e.g., a pH, a consumer perception) of the chem ical composition may have uniquevalues, for example, based on the ingredients w ithin the composition. The value of the property ofthe chemical composition may change due to the interaction of the ingredients within the composition.Produc ts may also have attributes in addition to its ingredients, such as sensory attr ibutes of the productand/or packaging of the product. Sensory attributes may in clude consistencies of the product, text of thepackaging of the product, and co lors of the product and packaging of the product. Such sensory attributesmay be influential in the purchase of the product, and, therefore, may result in the m arket value of theproduct.

    'Real-Time Ensemble Evaluation' in Patent Application Approval Process (USPTO 20 240232726)

    162-165页
    查看更多>>摘要:This patent application has not been assigned to a company or institution.The following quote was obtained by the news editors from the background informa tion supplied by theinventors: “The present invention relates generally to the field of machine learning, and more particularlyto mitigating model drift.“Machine learning models are susceptible to model drift, where existing models b ecome increasinglyineffective (i.e., model accuracy reductions) due to data cha nges over time as new data is incorporated.Traditional systems deploy one or mo re models and retrain said models once new data deviates (i.e., drifts) from the original training set, although detecting and determining drift is complex and computationally expensivefor many problem domains. For example, a financial for ecasting model that predicts next quarterlyrevenue cannot retrain until the fis cal quarter passes and actual revenue is observed and transformed intoassociate d labels/predictions. Models that cannot dynamically incorporate new data become outdatedand fail to generalize future data, decreasing the overall effectivene ss of the models and the system as awhole. Traditionally, as new data is incorp orated into new, retrained, models, said models degrade (i.e.,drift) due to the removal of relevant data in previous models, affecting the performance of the e nsemble.To avoid this drift, systems perform general classifier evaluation meas ures to evaluate model performancecorresponding to a specific historical period , but within the historical period evaluation measures arestatic, (i.e., insens itive to real-time status and conditions), and highly influenced by hyperparamet ers.This insensitivity to real-time events decreases ensemble/model accuracy an d generalizability.

    Patent Issued for Artificial intelligence character models with goaloriented be havior (USPTO 12033265)

    166-168页
    查看更多>>摘要:Patent number 12033265 is assigned to Theai Inc. (Mountain View, California, Uni ted States).The following quote was obtained by the news editors from the background informa tion supplied by theinventors: “Virtual characters are widely used in various s oftware applications, such as games, metaverses,social media, messengers, video communication tools, and online training tools. Some of these applicationsallo w users to interact with virtual characters. However, existing models of virtual characters are typicallydeveloped for specific applications and do not allow i ntegration with other applications and environments.Moreover, existing virtual character models are typically based on descriptions of specific rules and follo wspecific logic.