Abstract: A method of manufacturing a housing component for an aerosol generation device includes injection molding a housing component between first and second die parts. The first die part has a concave surface so as to form a convex surface on the component during the molding, the concave surface of the die having a first texture which imparts a corresponding texture on the convex surface of the component during molding, wherein the first texture includes a variable roughness across the concave surface of the first die part. Also provided are housing components for an aerosol generation device, the housing components having convex surfaces with textures.

Abstract: An eye mask with movable speakers is provided. The eye mask has a mask main body and two speaker assemblies. The mask main body is configured to cover a user's eyes and ears and comprises two inner spaces. The two inner spaces correspond to the user's ears in location. The two speaker assemblies are mounted in the inner spaces respectively, and at least one of the speaker assemblies is capable of moving in the corresponding inner space. When wearing the present eye mask, the user may cover eyes first and then connect and tie two ends of the mask main body, thereby securing the eye mask on the head. In the meantime, the mask also covers the user's ears. Then, the two speaker assemblies may be moved in the inner spaces to align with the user's ears respectively.

Abstract: An eye mask with movable speakers is provided. The eye mask has a mask main body and two speaker assemblies. The mask main body is configured to cover a user's eyes and ears and comprises two inner spaces. The two inner spaces correspond to the user's ears in location. The two speaker assemblies are mounted in the inner spaces respectively, and at least one of the speaker assemblies is capable of moving in the corresponding inner space. When wearing the present eye mask, the user may cover eyes first and then connect and tie two ends of the mask main body, thereby securing the eye mask on the head. In the meantime, the mask also covers the user's ears. Then, the two speaker assemblies may be moved in the inner spaces to align with the user's ears respectively.

Abstract: The general problem of pattern change discovery between high-dimensional data sets is addressed by considering the notion of the principal angles between the subspaces is introduced to measure the subspace difference between two high-dimensional data sets. Current methods either mainly focus on magnitude change detection of low-dimensional data sets or are under supervised frameworks. Principal angles bear a property to isolate subspace change from the magnitude change. To address the challenge of directly computing the principal angles, matrix factorization is used to serve as a statistical framework and develop the principle of the dominant subspace mapping to transfer the principal angle based detection to a matrix factorization problem. Matrix factorization can be naturally embedded into the likelihood ratio test based on the linear models. The method may be unsupervised and addresses the statistical significance of the pattern changes between high-dimensional data sets.

Abstract: An eye mask for sleeping has a headband and two eyecups. The headband comprises an inner surface and an outer surface. The two eyecups are configured to cover two eyes respectively. Each eyecup has an adhering surface, a contact surface, and multiple first passages. The adhering surface is securely mounted on the inner surface of the headband. The contact surface is configured to cover one of the eyes of the user. The first passages are formed through each eyecup and extend from the adhering surface to the contact surface. With the eyecups mounted on the headband and configured to cover the eyes of the user, because the first passages are formed through the eyecups, heat and moisture can be dissipated and drained via the first passages.

Abstract: An eye mask with movable speakers is provided. The eye mask has a mask main body and two speaker assemblies. The mask main body is configured to cover a user's eyes and ears and comprises two inner spaces. The two inner spaces correspond to the user's ears in location. The two speaker assemblies are mounted in the inner spaces respectively, and at least one of the speaker assemblies is capable of moving in the corresponding inner space. When wearing the present eye mask, the user may cover eyes first and then connect and tie two ends of the mask main body, thereby securing the eye mask on the head. In the meantime, the mask also covers the user's ears. Then, the two speaker assemblies may be moved in the inner spaces to align with the user's ears respectively.

Abstract: A multi-piece eye mask is provided and has a first headband, a second headband, and two speaker assemblies. The first headband is configured to cover the user's eyes and ears. The second headband is detachably mounted on the first headband. The speaker assemblies are mounted in the second headband and corresponding to the user's ears in location. The first headband can be separated from the second headband, which contains the eye cups, and the first headband with the eye cups is independently washable.

Abstract: The general problem of pattern change discovery between high-dimensional data sets is addressed by considering the notion of the principal angles between the subspaces is introduced to measure the subspace difference between two high-dimensional data sets. Current methods either mainly focus on magnitude change detection of low-dimensional data sets or are under supervised frameworks. Principal angles bear a property to isolate subspace change from the magnitude change. To address the challenge of directly computing the principal angles, matrix factorization is used to serve as a statistical framework and develop the principle of the dominant subspace mapping to transfer the principal angle based detection to a matrix factorization problem. Matrix factorization can be naturally embedded into the likelihood ratio test based on the linear models. The method may be unsupervised and addresses the statistical significance of the pattern changes between high-dimensional data sets.

Abstract: A test socket comprising a guide plate with a lower surface engaged with an upper surface of a main test structure, the guide plate further including an upper surface which is parallel to the lower surface and an opening extending through the guide plate, the main test structure includes a body with one or more apertures through the upper surface and one or more probes mounted within the main test structure, the probes including a front end which extends through the apertures for engagement by a lead or terminal pad of a device to be tested, and a tail end which is secured within the main test structure by an elastomeric material.