Abstract: Disclosed solutions for improved machine learning (ML) employ knowledge balancing self-distillation with adaptive mutual information (AMI). Examples include: for a neural network (NN) having a plurality of modules, determining a task objective for at least a final module of the plurality of modules; for the NN, determining a balancing objective using at least an output of the final module and an output of a first intermediate module of the plurality of modules; determining an overall objective, wherein determining the overall objective comprises combining the task objective with the balancing objective; and adjusting weights of the NN to minimize the overall objective. Balancing information may combine mutual information (between an intermediate module output and the output of the final module) with self-information (for the intermediate module output) to produce AMI. Adjusting weights of the NN during training, using the AMI, results in knowledge balancing self-distillation.

Abstract: A classifier model is trained for temporal action localization of video clips. A training video clip that includes actions of interest for identification is ingested into the classifier model. Action characteristics within frames of the video clip are identified. The actions correspond to known action classes. An actionness score is determined for each of the frames based upon the action characteristics identified within each of the frames. Class activation sequence (CAS) scores are determined for sequences of the frames based upon a presence or an absence of the action characteristics identified within each of the frames. Base confidence predictions of temporal locations of actions of interest within the video clip are produced by correlating each of the actionness scores with corresponding class activation scores for each of the frames in the sequences of frames.

Abstract: A neural architecture search (NAS) with a weak predictor comprises: receiving network architecture scoring information; iteratively sampling a search space, wherein the sampling comprises: generating a set of candidate architectures within the search space; learning a first predictor; evaluating performance of the candidate architectures; and based on at least the performance of the set of candidate architectures and the network architecture scoring information, refining the search space to a smaller search space; based on at least the network architecture scoring information, thresholding the performance of candidate architectures to determine scored output candidate architectures; and reporting the scored output candidate architectures. In some examples, the candidate architectures each comprise a machine learning (ML) model, for example a neural network (NN).

Abstract: This disclosure relates to methods for treating cancer in a subject identified as having chronic myelomonocytic leukemia (CMML), comprising administering an anti-ILT3 antigen binding protein, or antigen binding fragment to the patient every three weeks (Q3W).

Abstract: Example solutions for video object segmentation (VOS) use a bilateral attention transformer in motion-appearance neighboring space, and perform a process that includes: receiving a video stream comprising a plurality of video frames in a sequence; receiving a first object mask for an initial video frame of the plurality of video frames; selecting a video frame of the plurality of video frames as a current query frame, the current query frame following, in the sequence, a reference frame of a reference frame set, wherein each reference frame has a corresponding object mask; using the current query frame and a video frame in the reference frame set, determining a bilateral attention; and using the bilateral attention, generating an object mask for the current query frame.

Abstract: The present application provides a display panel, a fabrication method of a display panel, and a display device. The display panel includes a transparent display panel, and a dimming panel comprising a plurality of pixels for emitting light and is arranged on a side of the transparent display panel. The dimming panel includes an electrochromic layer, an electrolyte layer, and an ion storage layer stacked in sequence. The electrochromic layer and the ion storage layer includes a porous structure, and the dimming panel further includes a first transparent electrode layer disposed on a side of the electrochromic layer away from the electrolyte layer, and the first transparent electrode layer is patterned and disposed corresponding to each pixel.

Abstract: Some embodiments describe a DC meter system that can be placed in-line between a power source and an electrical load. In some embodiments, the DC meter system includes a DC meter that has multiple inputs for receiving electricity from multiple power sources. In some embodiments, the DC meter has one or more outputs, each capable of delivering electricity to a different electrical load. In some embodiments, the DC meter can combine electricity from two or more of the multiple inputs into one of the outputs. In some embodiments, the DC meter has a controller that can enable a user to deliver electricity from a particular input to a particular output. In some embodiments, the DC meter is configured to enable transition of electricity between an input and an output in a bi-directional manner. In some embodiments, the DC meter can record an amount of electricity transmitted in either direction.

Abstract: The invention described herein includes a self-neutralizing amino acid-based cationic composition that includes: from 10 wt % to 70 wt % of an amino acid-based cationic ester of formula (I): where R1 is a branched or linear alkyl group having 1 to 10 carbon atoms, R2 is a linear or branched carbon chain having 10 to 50 carbon atoms, where X? represents a conjugate base; from 2 wt % to 25 wt % of an anhydrous buffering agent comprising a salt formed from a strong base and a weak organic acid; and a nonionic amphiphile having 10 to 35 carbon atoms in an amount sufficient to total 100 wt % of the composition. The composition may be in anhydrous solid form and may be a dispersion of particles of the anhydrous buffering agent in a mixture of the amino acid-based cationic ester and the nonionic amphiphile.

Abstract: A method of balancing a dataset for a machine learning model includes identifying confusing classes of few-shot classes for a machine learning model during validation. One of the confusing classes and an image from one of the few-shot classes are selected. An image perturbation is computed such that the selected image is classified as the selected confusing class. The selected image is modified with the computed perturbation. The modified selected image is added to a batch for training the machine learning model.

Abstract: The invention described herein includes a self-neutralizing amino acid-based cationic composition that contains an amino acid-based cationic ester, a nonionic amphiphile, and an anhydrous buffering agent. In an embodiment, the composition, when dispersed or dissolved in an aqueous solvent, provides a resultant solution/dispersion having a pH of greater than about 4. The amino acid-based cationic ester may be a reaction product of (i) an amino acid having a non-polar side chain, and an amine group that has been neutralized with an acid; with (ii) a long chain fatty alcohol. Also included within the scope of the invention are methods of preparing a self-neutralizing amino acid-based cationic composition, such method including combining an amino acid-based cationic ester, a nonionic amphiphile and an anhydrous buffering agent. The resulting composition is in an anhydrous solid form. Formulations containing the amino acid-based cationic ester and/or a composition that includes it, are also disclosed.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for privacy-sensitive training of a neural network. In one aspect, a system comprises a central memory configured to store current values of a set of neural network parameters and one or more computers that are configured to implement a plurality of worker computing units, where each worker computing unit is configured to repeatedly perform operations comprising obtaining current values of the set of neural network parameters from the central memory, sampling a batch of network inputs from a set of training data, determining a respective gradient corresponding to each network input, determining an aggregated gradient based on the gradients, identifying a subset of a set of gradient values as target values, generating a noisy gradient by combining random noise with the target gradient values, and updating the current values of the set of neural network parameters.

Abstract: A computer device for automatic feature detection comprises a processor, a communication device, and a memory configured to hold instructions executable by the processor to instantiate a dynamic convolution neural network, receive input data via the communication network, and execute the dynamic convolution neural network to automatically detect features in the input data. The dynamic convolution neural network compresses the input data from an input space having a dimensionality equal to a predetermined number of channels into an intermediate space having a dimensionality less than the number of channels. The dynamic convolution neural network dynamically fuses the channels into an intermediate representation within the intermediate space and expands the intermediate representation from the intermediate space to an expanded representation in an output space having a higher dimensionality than the dimensionality of the intermediate space.

Abstract: Example solutions for video frame action detection use a gated history and include: receiving a video stream comprising a plurality of video frames; grouping the plurality of video frames into a set of present video frames and a set of historical video frames, the set of present video frames comprising a current video frame; determining a set of attention weights for the set of historical video frames, the set of attention weights indicating how informative a video frame is for predicting action in the current video frame; weighting the set of historical video frames with the set of attention weights to produce a set of weighted historical video frames; and based on at least the set of weighted historical video frames and the set of present video frames, generating an action prediction for the current video frame.

Abstract: The present application provides a display panel, a fabrication method of a display panel, and a display device. The display panel includes a transparent display panel, and a dimming panel comprising a plurality of pixels for emitting light and is arranged on a side of the transparent display panel. The dimming panel includes an electrochromic layer, an electrolyte layer, and an ion storage layer stacked in sequence. The electrochromic layer and the ion storage layer includes a porous structure, and the dimming panel further includes a first transparent electrode layer disposed on a side of the electrochromic layer away from the electrolyte layer, and the first transparent electrode layer is patterned and disposed corresponding to each pixel.

Abstract: A classifier model is trained for temporal action localization of video clips. A training video clip that includes actions of interest for identification is ingested into the classifier model. Action characteristics within frames of the video clip are identified. The actions correspond to known action classes. An actionness score is determined for each of the frames based upon the action characteristics identified within each of the frames. Class activation sequence (CAS) scores are determined for sequences of the frames based upon a presence or an absence of the action characteristics identified within each of the frames. Base confidence predictions of temporal locations of actions of interest within the video clip are produced by correlating each of the actionness scores with corresponding class activation scores for each of the frames in the sequences of frames.

Abstract: Some embodiments describe a DC meter system that can be placed in-line between a power source and an electrical load. In some embodiments, the DC meter system includes a DC meter that has multiple inputs for receiving electricity from multiple power sources. In some embodiments, the DC meter has one or more outputs, each capable of delivering electricity to a different electrical load. In some embodiments, the DC meter can combine electricity from two or more of the multiple inputs into one of the outputs. In some embodiments, the DC meter has a controller that can enable a user to deliver electricity from a particular input to a particular output. In some embodiments, the DC meter is configured to enable transition of electricity between an input and an output in a bi-directional manner. In some embodiments, the DC meter can record an amount of electricity transmitted in either direction.

Abstract: This disclosure relates to methods for treating solid tumors in a patient identified as having metastatic triple negative breast cancer, glioblastoma, metastatic pancreatic ductal adenocarcinoma, metastatic soft tissue sarcoma, or metastatic non-squamous non-small cell lung carcinoma, comprising administering an anti-ILT3 antigen binding protein, or antigen binding fragment, and an anti-PD1 antigen binding protein, or antigen binding fragment, to the patient every three weeks (Q3W).

Abstract: The disclosure herein describes preparing and using a cross-attention model for action recognition using pre-trained encoders and novel class fine-tuning. Training video data is transformed into augmented training video segments, which are used to train an appearance encoder and an action encoder. The appearance encoder is trained to encode video segments based on spatial semantics and the action encoder is trained to encode video segments based on spatio-temporal semantics. A set of hard-mined training episodes are generated using the trained encoders. The cross-attention module is then trained for action-appearance aligned classification using the hard-mined training episodes. Then, support video segments are obtained, wherein each support video segment is associated with video classes. The cross-attention module is fine-tuned using the obtained support video segments and the associated video classes.

Abstract: A computerized method trains and uses a multimodal fusion transformer (MFT) model for content moderation. Language modality data and vision modality data associated with a multimodal media source is received. Language embeddings are generated from the language modality data and vision embeddings are generated from the vision modality data. Both kinds of embeddings are generated using operations and/or processes that are specific to the associated modalities. The language embeddings and vision embeddings are combined into combined embeddings and the MFT model is used with those combined embeddings to generate a language semantic output token, a vision semantic output token, and a combined semantic output token. Contrastive loss data is generated using the three semantic output tokens and the MFT model is adjusted using that contrastive loss data. After the MFT model is trained sufficiently, it is configured to perform content moderation operations using semantic output tokens.