Abstract: Continual learning techniques are described for extending the capabilities of a base model, which is trained to predict a set of existing or base classes, to generate a target model that is capable of making predictions for both the existing or base classes and additionally for making predictions for new or custom classes. The techniques described herein enable the target model to be trained such that the model can make predictions involving both base classes and custom classes with high levels of accuracy.

Abstract: A device captures a series of images over time in association with a gate, each image having a timestamp. The device determines, for a vehicle approaching the entry side, from a subset of images of the series of images featuring the vehicle, a first data set comprising a plurality of parameters that describe attributes of the vehicle by inputting the subset of images into a first machine learning model and a vehicle identifier of the vehicle by inputting images of the subset featuring a depiction of a license plate of the vehicle into a second machine learning model. The device stores the data set in association with one or more timestamps with the subset of images, determines a second data set for a second vehicle approaching the exit side, and responsive to determining that the first data set and the second data set match, instructs the gate to move.

Abstract: Artificial intelligence generated text detector includes performing operations that include obtaining an input document, parsing the input document into tokens, generating, by an embedding model, token embeddings for the tokens, and aggregating the token embeddings across the input document to obtain a document embedding. The operations further include performing, by a document prediction model processing the document embedding, a document level classification to obtain a set of document class probabilities including an AI generated probability, a mixed probability, and a human generated probability. The operations further include presenting the set of document class probabilities.

Abstract: Techniques are described herein for generating block extender model. An example method includes a system accessing a base model trained for identifying a base class. The system can access an extender comprising block extenders, the extender class distinct from the base class. The system can connect the extender with the base model to generate an augmented model. The system can input training data to the augmented model, the training data is provided to the base model and the extender, the training data comprising data points associated with the extender class. The system can train the extender model to identify the extender class based at least in part on the training data and the signal received from the base machine learning model. The system can generate a trained extender based at least in part on the training, the extender trained to identify an object associated with the extender class.

Abstract: A device captures a series of images over time in association with a gate, each image having a timestamp. The device determines, for a vehicle approaching the entry side, from a subset of images of the series of images featuring the vehicle, a first data set comprising a plurality of parameters that describe attributes of the vehicle by inputting the subset of images into a first machine learning model and a vehicle identifier of the vehicle by inputting images of the subset featuring a depiction of a license plate of the vehicle into a second machine learning model. The device stores the data set in association with one or more timestamps with the subset of images, determines a second data set for a second vehicle approaching the exit side, and responsive to determining that the first data set and the second data set match, instructs the gate to move.

Abstract: A device captures a series of images over time in association with a gate, each image having a timestamp. The device determines, for a vehicle approaching the entry side, from a subset of images of the series of images featuring the vehicle, a first data set comprising a plurality of parameters that describe attributes of the vehicle by inputting the subset of images into a first machine learning model and a vehicle identifier of the vehicle by inputting images of the subset featuring a depiction of a license plate of the vehicle into a second machine learning model. The device stores the data set in association with one or more timestamps with the subset of images, determines a second data set for a second vehicle approaching the exit side, and responsive to determining that the first data set and the second data set match, instructs the gate to move.

Abstract: Continual learning techniques are described for extending the capabilities of a base model, which is trained to predict a set of existing or base classes, to generate a target model that is capable of making predictions for both the existing or base classes and additionally for making predictions for new or custom classes. The techniques described herein enable the target model to be trained such that the model can make predictions involving both base classes and custom classes with high levels of accuracy.

Abstract: Methods are provided for producing a shrunken replica or a shrunken inverse replica of a multicontinuous structure. In one embodiment, the method comprises (1) infiltrating a first multicontinuous structure with a first fluid material which can be immobilized and shrunken, wherein the first multicontinuous structure comprises at least a first phase and a second phase which is immiscible with the first phase, and the infiltration displaces the second phase in the first multicontinuous structure to form a second multicontinuous structure which comprises at least the first phase and a third phase which consists of the first fluid material; (2) immobilizing the infiltrated first fluid material in the second multicontinuous structure; (3) removing the first phase from the second multicontinuous structure; and (4) shrinking (e.g., by a pyrolysis process) the third phase of the second multicontinuous structure, to produce an isotropically shrunken inverse replica of the first multicontinuous structure.

Abstract: A structured material is disclosed with magneto-gyrotropic characteristics including at least one continuous structurally-chiral material. The structured material has an electric permittivity and a magnetic permeability at least one of which varies within the structured material along a first direction in a repetitious fashion wherein a repetition unit includes a chiral component and is at least 25 nm in length. The structured material exhibits non-reciprocal electromagnetic wave propagation velocity characteristics along a second direction that includes a non-zero component along the first direction.

Abstract: A device having at least one dielectric inner core region in which electromagnetic radiation is confined, and at least two dielectric outer regions surrounding the inner core region, each with a distinct refractive index. The outer regions confine electromagnetic radiation within the inner core region. The refractive indices, the number of outer regions, and thickness of the outer regions result in a reflectivity for a planar geometry that is greater than 95% for angles of incidence ranging from 0° to at least 80° for all polarizations for a range of wavelengths of the electromagnetic radiation. In exemplary embodiments, the inner core region is made of a low dielectric material, and the outer regions include alternating layers of low and high dielectric materials. In one aspect of the invention, the device is a waveguide, and in another aspect the device is a microcavity.

Abstract: An all-dielectric coaxial waveguide comprising a dielectric core region; an annulus of dielectric material, surrounding the core region, in which electromagnetic radiation is confined; and an outer region of cylindrically coaxial dielectric shells of alternating indices of refraction surrounding the annulus. The core region and the outer region have an average index of refraction which is higher than the index of refraction of the annulus.

Abstract: A device having at least one dielectric inner core region in which electromagnetic radiation is confined, and at least two dielectric outer regions surrounding the inner core region, each with a distinct refractive index. The outer regions confine electromagnetic radiation within the inner core region. The refractive indices, the number of outer regions, and thickness of the outer regions result in a reflectivity for a planar geometry that is greater than 95% for angles of incidence ranging from 0° to at least 80° for all polarizations for a range of wavelengths of the electromagnetic radiation. In exemplary embodiments, the inner core region is made of a low dielectric material, and the outer regions include alternating layers of low and high dielectric materials. In one aspect of the invention, the device is a waveguide, and in another aspect the device is a microcavity.

Abstract: A device having at least one dielectric inner core region in which electromagnetic radiation is confined, and at least two dielectric outer regions surrounding the inner core region, each with a distinct refractive index. The outer regions confine electromagnetic radiation within the inner core region. The refractive indices, the number of outer regions, and thickness of the outer regions result in a reflectivity for a planar geometry that is greater than 95% for angles of incidence ranging from 0° to at least 80° for all polarizations for a range of wavelengths of the electromagnetic radiation. In exemplary embodiments, the inner core region is made of a low dielectric material, and the outer regions include alternating layers of low and high dielectric materials. In one aspect of the invention, the device is a waveguide, and in another aspect the device is a microcavity.

Abstract: The present invention relates to the field of redefining absolute references in a Web site so as to make the Web site and its referenced content portable and installing the portable Web site on a system within the mezzanine of the Internet so as to make the information available to a user from a more proximate location. To take advantage of the unused capacity of mezzanine systems, the present invention incorporates a systematic, organized, and highly structured, yet customizable, procedure for optimizing deployment of data from a remote primary server to any platform or location from which the data would be most optimally accessed. This is accomplished by globally replacing Web site source references with tokens. The tokens are then instantiated on install time and resolved specific to the target platform.