Abstract: Methods are presented for providing dynamic search filter suggestions that are updated and ranked based on the user filter selections. One method includes detecting a query received in a user interface (UI), calculating, by a search-candidate model, first search results, and calculating, by a suggestions model, first filter suggestions for filter categories to filter responses to the query. The suggestions model is obtained by training a machine-learning algorithm utilizing pairwise learning-to-rank modeling. The first search results and the first filter suggestions are presented in the UI. When a selection in the UI of a filter suggestion is detected, the search-candidate model calculates second search results for the filter categories based on the query and the selected filter suggestion, and the suggestions model calculates second first filter suggestions based on the query and the selected filter suggestion. The second search results and the second filter suggestions are presented in the UI.

Abstract: Provided are structurally-reconfigurable, optical metasurfaces constructed by, for example, integrating a plasmonic lattice array in the gap between a pair of microbodies that serve to locally amplify the strain created on an elastomeric substrate by an external mechanical stimulus. The spatial arrangement and therefore the optical response of the plasmonic lattice array is reversible.

Abstract: An organic-electrolyte lithium-oxygen battery with a full-enclosed structure and a preparation method thereof are disclosed. In the present disclosure, a lithium-oxygen battery unit is enclosed in a shell containing pure oxygen, and the reactant oxygen is recycled without additional supply. Among them, a part of oxygen is stored in the form of lithium peroxide by pre-discharging. When in use, a charging is firstly performed to decompose the lithium peroxide to release the fixed oxygen.

Abstract: The disclosed technologies include a navigation agent for a search interface. In an embodiment, the navigation agent uses reinforcement learning to dynamically generate and select navigation options for presentation to a user during a search session. The navigation agent selects navigation options based on reward scores, which are computed using implicit and/or explicit user feedback received in response to presentations of navigation options.

Abstract: Methods are presented for providing dynamic search filter suggestions that are updated and ranked based on the user filter selections. One method includes detecting a query received in a user interface (UI), calculating, by a search-candidate model, first search results, and calculating, by a suggestions model, first filter suggestions for filter categories to filter responses to the query. The suggestions model is obtained by training a machine-learning algorithm utilizing pairwise learning-to-rank modeling. The first search results and the first filter suggestions are presented in the UI. When a selection in the UI of a filter suggestion is detected, the search-candidate model calculates second search results for the filter categories based on the query and the selected filter suggestion, and the suggestions model calculates second first filter suggestions based on the query and the selected filter suggestion. The second search results and the second filter suggestions are presented in the UI.

Abstract: In an example embodiment, position bias and other types of bias may be compensated for by using two-phase training of a machine-learned model. In a first phase, the machine-learned model is trained using non-randomized training data. Since certain types of machine-learned models, such as those involving deep learning (e.g., neural networks) require a lot of training data, this allows the bulk of the training to be devoted to training using non-randomized training data. However, since this non-randomized training data may be biased, a second training phase is then used to revise the machine-learned model based on randomized training data to remove the bias from the machine-learned model. Since this randomized training data may be less plentiful, this allows the deep learning machine-learned model to be trained to operate in an unbiased manner without the need to generate additional randomized training data.

Abstract: Provided are structurally-reconfigurable, optical metasurfaces constructed by, for example, integrating a plasmonic lattice array in the gap between a pair of microbodies that serve to locally amplify the strain created on an elastomeric substrate by an external mechanical stimulus. The spatial arrangement and therefore the optical response of the plasmonic lattice array is reversible.

Abstract: Techniques for suggesting targeting criteria for a content delivery campaign are provided. An affinity score representing an affinity between the attribute values of each pair of multiple pairs of attribute values is computed. First input indicating a particular attribute value for a particular attribute type is received through a user interface for creating a content delivery campaign. The user interface includes fields for inputting attribute values for multiple attribute types that includes the particular attribute type. In response to the first input and based on affinity scores associated with the particular attribute value, a set of suggested attribute values is identified. The user interface is updated to include the set of suggested attribute values. Second input indicating a selection of a particular suggested attribute value is received. The particular suggested attribute value is added to the content delivery campaign.

Abstract: In an example embodiment, position bias and other types of bias may be compensated for by using two-phase training of a machine-learned model. In a first phase, the machine-learned model is trained using non-randomized training data. Since certain types of machine-learned models, such as those involving deep learning (e.g., neural networks) require a lot of training data, this allows the bulk of the training to be devoted to training using non-randomized training data. However, since this non-randomized training data may be biased, a second training phase is then used to revise the machine-learned model based on randomized training data to remove the bias from the machine-learned model. Since this randomized training data may be less plentiful, this allows the deep learning machine-learned model to be trained to operate in an unbiased manner without the need to generate additional randomized training data.

Abstract: Disclosed herein are systems, methods, and non-transitory computer-readable media for providing user notifications based on a project context. The system may receive candidate attributes from candidate devices of a plurality of candidates and storing them in a candidate database, as well as user-entered attributes from a user device of a user. The system may then iteratively execute a number of operations that include performing a search for candidates in the candidate database by comparing project attributes with candidate attributes and providing user notification of newly-matched candidates that includes returning returned candidates that are matching candidates of the search results to the user based on the search.