Abstract: A curable composition includes an epoxy monomer component and an aniline-based hardener. The epoxy monomer component is a first component formed from a first epoxy monomer represented by Formula (I), or a second component including the first epoxy monomer represented by Formula (I) and a second epoxy monomer different from the first epoxy monomer represented by Formula (I), wherein each of the substituents in Formula (I) is given the definitions as set forth in the Specification and Claims. Based on 100 wt % of the epoxy monomer component, an amount of the first epoxy monomer represented by Formula (I) is not smaller than 25 wt % and less than 100 wt % and an amount of the second epoxy monomer is greater than 0% and not greater than 75 wt %. A cured product formed from the curable composition, and a method for encapsulating a semiconductor device using the curable composition are also provided.

Abstract: An operating method of a fully homomorphic encrypted neural network model is provided, wherein the fully homomorphic encrypted neural network model includes a plurality of layers, and the method performed by a processor includes: for one of the plurality of layers, encrypting a plaintext input with a first encryption algorithm to generate a ciphertext vector, performing a convolution operation according to the ciphertext vector to generate a result vector, transforming the result vector into a plurality of result ciphertexts adopting a second encryption algorithm, inputting the plurality of result ciphertexts into an activation function to generate a plurality of encrypted activation values, and repacking the plurality of encrypted activation values to generate an output vector adopting the first encryption algorithm.

Abstract: A machine learning training device is disclosed. The machine learning training device includes a virtual hard anchor generation circuit, a classification circuit and a training circuit. The virtual hard anchor generation circuit is configured to generate several virtual hard anchors according to several easy samples classified into several types. The virtual hard anchors respectively correspond to one of the several types. The classification circuit is configured to classify several hard samples into several types according to virtual hard anchors. Parts of the hard samples classified into several types are several clean hard samples. Another parts of the hard samples that are not classified into several types are several noisy hard samples. The training circuit is configured to perform machine learning training according to several easy samples and several clean hard samples.

Abstract: A Chinese herbal medicine extract, a method for preparing the same, and a use of the same are disclosed. The Chinese herbal medicine extract includes an active ingredient, which contains any one of or any combination of agarwood, Chinese honeylocust fruit, Chinese honeylocust spine, cinnamon leaf, and camphor leaf. Medication based on the Chinese herbal medicine extract is useful in treating coronavirus-related symptoms or diseases.

Abstract: An inference method for encrypted deep neural network model is executed by a computing device and includes: encoding a message according to a quantization parameter to generate a plaintext, encrypting the plaintext according to a private key to generate a ciphertext, sending the ciphertext to a deep neural network model to generate a ciphertext result, decrypting the ciphertext result according to the private key to generate a plaintext result, and decoding the plaintext result according to the quantization parameter to generate an inference result.

Abstract: Provided is a semiconductor device including: a substrate, a plurality of isolation structures, a plurality of channel layers, and a gate structure. The substrate includes a plurality of fins thereon. The plurality of isolation structures are respectively disposed between the plurality of fins. A top surface of the plurality of isolation structures is higher than a top surface of the plurality of fins to form a plurality of openings. The plurality of channel layers are respectively disposed in the plurality of openings. Each channel layer is in contact with a corresponding fin and extends to cover a lower sidewall of a corresponding isolation structure, thereby forming a U-shaped structure. The gate structure is filled in the plurality of openings and extends to cover the top surface of the plurality of isolation structures.

Abstract: An elastic glasses temple includes a connector, a temple body having a round hook in a front end to pivotally connected to the connector, a sleeve fitting around a front portion of the temple body with a coil spring fitted in a second hole formed in a rear portion of the sleeve, a coil spring and a position member combined together. The position member is fixed on the temple body, fitting in a rear end of the sleeve. When the temple body is forced to swing outward by exterior force, with a screw fitting in said round hook as a pivot, the sleeve is pushed to move rearward to compress the coil spring to absorb exterior force to permit the temple body swing outward. The temple body swings back after exterior force disappears with resilience of the the coil spring. Then this temple can suit to any sizes of users' faces.

Abstract: An elastic temple for eyeglasses includes a temple body, a sleeve, a compression spring and a fixing member. The temple body has a front combine member to be pivotally combined with a lens frame, fitting through the sleeve and the compression spring and the fixing member. The compression spring is contained in a rear passageway of the sleeve, and the fixing member pushes the rear end of the spring and is fixed firmly on the temple body. The sleeve can be pushed by the spring and move between the front combine member and the fixing member so that the temple body may be swung outward and back to its original position, with a screw connecting the lens frame and the elastic temple functioning as a pivot.

Abstract: The present invention relates to a vehicular rearview mirror, particularly a vehicular rearview mirror which can be electrically adjusted by a driver sitting in the driver's seat by pressing lightly a toggle switch with one hand, enabling the rearview mirror to shift its direction upward or downward, to the left or to the right and arrive at the most suitable reflection angle.