Deep Interest Network (DIN) is a state-of-the-art model which uses attention mechanism to capture user interests from historical behaviors. User interests intuitively follow a hierarchical pattern such that users generally show interests from a higher-level then to a lower-level abstraction. Modelling such interest hierarchy in an attention network can fundamentally improve the representation of user behaviors. We therefore propose an improvement over DIN to model arbitrary interest hierarchy: Deep Interest with Hierarchical Attention Network (DHAN). In this model, a multi-dimensional hierarchical structure is introduced on the first attention layer which attends to individual item, and the subsequent attention layers in the same dimension attend to higher-level hierarchy built on top of the lower corresponding layers. To enable modelling of multiple dimensional hierarchy, an expanding mechanism is introduced to capture one to many hierarchies. This design enables DHAN to attend different importance to different hierarchical abstractions thus can fully capture a user's interests at different dimensions (e.g. category, price or brand). To validate our model, a simplified DHAN is applied to Click-Through Rate (CTR) prediction and our experimental results on three public datasets with two levels of one-dimensional hierarchy only by category. It shows DHAN's superiority with significant AUC uplift from 12% to 21% over DIN. DHAN is also compared with another state-of-the-art model Deep Interest Evolution Network (DIEN), which models temporal interest. The simplified DHAN also gets slight AUC uplift from 1.0% to 1.7% over DIEN. A potential future work can be combination of DHAN and DIEN to model both temporal and hierarchical interests.

An effective method to restore images blurred by uniform linear motion is presented. The effect of the out-of-frame pixels on the blurring process and the estimate of these pixels are analyzed. The restoration qualities of different deblurring methods are compared. It is concluded that it is impossible to determine the length of blurring movement in the frequency domain.>

A principle of measuring the area of small kinds of objects is proposed which a method of simulating the procedure of subjective visual assessment of yarn appearance is established using a microcomputer vision system. An algorithm was designed to make data compression and shading corrections, and to construct a matrix containing yarn thickness information with enough accuracy. Using the dependence method, a cooccurrence matrix is constructed from which two parameters are drawn. The two parameters conform well with the subjective conclusions.>

A hierarchical relaxation labeling algorithm incorporating a decimation-in-time Hough transform is proposed. Relaxation clues are star structure and circuits. Existing techniques for registration of stereo vision are first reviewed and synthesized. Feature extraction, including a decimation-in-time Hough transform, star structured and relation, and circuit coding using a Freeman code, is carried out. The hierarchical relaxation labeling procedure and the fully constrained compatibility coefficients are then treated.>