Abstract: The present disclosure relates to a new method for cross-modal retrieval via deep binary hashing and quantization. In a training phase, the system simultaneously learns to generate feature vectors, binary codes, and quantization codes for data across two or more modalities that preserves the semantic similarity correlations in the original data. In a prediction phase, the system retrieve a data item in a database that is semantically similar to a query item of a different modality. To identify the database item closest in semantic meaning to the query item, the system first narrows the database search space based on binary hash code distances between each of the database items and the query item. The system then measures the quantization distances between the query items and the database items in the smaller search space. The system identifies database item have the closest quantization distance to the query item as the closest semantic match to the query item.

Abstract: The present disclosure relates to improving recommendations for small shops on an ecommerce platform while maintaining accuracy for larger shops. The improvement is achieved by retraining a machine-learning recommendation model to reduce sample selection bias using a meta-learning process. The retraining process comprises identifying a sample subset of shops on the ecommerce platform, and then creating shop-specific versions of the recommendation model for each of the shops in the subset. Each shop-specific model is created by optimizing the baseline model to predict user-item interactions in a first training dataset for the applicable shop. Each of the shop-specific models is then tested using a second training dataset for the shop. A loss is calculated for each shop-specific model based on the model's predicted user-item interactions and the actual user-item interactions in the second training dataset for the shop.

Abstract: The present disclosure relates to a new method for cross-modal retrieval via deep binary hashing and quantization. In a training phase, the system simultaneously learns to generate feature vectors, binary codes, and quantization codes for data across two or more modalities that preserves the semantic similarity correlations in the original data. In a prediction phase, the system retrieve a data item in a database that is semantically similar to a query item of a different modality. To identify the database item closest in semantic meaning to the query item, the system first narrows the database search space based on binary hash code distances between each of the database items and the query item. The system then measures the quantization distances between the query items and the database items in the smaller search space. The system identifies database item have the closest quantization distance to the query item as the closest semantic match to the query item.

Abstract: An optical fiber for maximizing residual mechanical stress and an optical fiber grating fabricating method using the optical fiber are provided. The optical fiber includes a core formed of silica, for propagating light, and a cladding formed of boron-doped silica, surrounding the core. Alternatively, the optical fiber includes a core formed of phosphorous-doped silica and a cladding formed of silica, surrounding the core.

Abstract: An optical fiber for maximizing residual mechanical stress and an optical fiber grating fabricating method using the optical fiber are provided. The optical fiber includes a core formed of silica, for propagating light, and a cladding formed of boron-doped silica, surrounding the core. Alternatively, the optical fiber includes a core formed of phosphorous-doped silica and a cladding formed of silica, surrounding the core.

Abstract: An optical fiber for maximizing residual mechanical stress and an optical fiber grating fabricating method using the optical fiber are provided. The optical fiber includes a core formed of silica, for propagating light, and a cladding formed of boron-doped silica, surrounding the core. Alternatively, the optical fiber includes a core formed of phosphorous-doped silica and a cladding formed of silica, surrounding the core.

Abstract: The present invention relates to a method of metal coating for an optical fiber grating with a long period. More specifically, the present invention relates to a method of using silver (Ag) as a coating material for an optical fiber through ascertaining its characteristics after silver is coated on the surface of the optical fiber using a silver mirror reaction.

Abstract: An erbium-doped optical fiber (EDF) and a fabricating method thereof The erbium-doped optical fiber has a core formed by substantially doping silica with erbium and having gratings formed therein at a predetermined period, for propagating light therethrough, and a cladding surrounding the core and having a lower refractive index than the core. Since the erbium-doped optical fiber acts as a gain flattening filter, an erbium-doped fiber amplifier formed out of the erbium-doped optical fiber obviates the need of splicing an erbium-doped optical fiber with a gain flattening filter.

Abstract: There is provided a device and method for forming optical fiber anti-symmetric long period gratings by microbending. In the optical fiber grating forming device, a laser system emits a CO2 laser beam, a lens focuses the CO2 laser beam on an optical fiber in a predetermined width, an optical fiber support fixes both ends of the optical fiber and imparts tensile strain to the optical fiber, and a controller controls the intensity of the CO2 laser beam, the focusing distance of the lens, and the tensile force applied by the optical fiber support.

Abstract: There are provided an optical fiber rod overcladding apparatus and method, and an optical fiber drawing method. In the preform rod overcladding method, a preform rod is clamped in a top chuck and leveled, and a glass tube is mounted in a bottom chuck and leveled. The preform rod is coaxially inserted into a glass tube. Then, the glass tube is preheated by the furnace and heated by the burner until the glass tube reaches a softening point. The preform rod is completely sealed in the glass tube by sucking air in a clearance between the preform rod and the glass tube by application of a negative vacuum pressure. Thus, a preform is completed.