Abstract: A three-dimensional semiconductor device may include a back-side metal layer, a lower channel pattern on the back-side metal layer, first and second lower source/drain patterns, which are spaced apart from each other in a first direction with the lower channel pattern interposed therebetween, the first lower source/drain pattern being connected to the lower channel pattern, an upper channel pattern on the lower channel pattern, a first upper source/drain pattern on the first lower source/drain pattern, the first upper source/drain pattern being connected to the upper channel pattern, a second upper source/drain pattern on the second lower source/drain pattern, and a wide via electrically connecting the first upper source/drain pattern to the second lower source/drain pattern. The wide via may include first and second via portions having first and second top surfaces, and here, the second top surface may be located at a level lower than the first top surface.

Abstract: Disclosed herein are a computing apparatus and method for performing reinforcement learning using a multimodal artificial intelligence agent. The method for performing reinforcement learning using a multimodal artificial intelligence agent includes: dividing frames, included in images acquired by capturing a virtual environment, into a plurality of sections; and performing reinforcement learning by applying any one of a plurality of guidance types to each of the plurality of sections and then allowing a multimodal artificial intelligence agent to interact with the virtual environment through the images. The plurality of guidance types is classified into three or more types according to their guidance level. Performing the reinforcement learning is performing reinforcement learning by applying a moderate-level guidance type to the sections of predetermined critical periods and also applying any one of the plurality of guidance types to the other sections.

Abstract: A toddler-inspired Bayesian learning method according to an embodiment includes: collecting information related to at least one task while an agent is performing exploration; and performing Bayesian inference regarding the at least one task by using the collected information as Bayesian informative priors.

Abstract: A wireless transmitting and receiving device includes antennas, weight adjusters connected to antennas, respectively, and configured to apply weights to an amplitude and a phase of a signal communicating through the antennas, signal transmitting and detecting units connected to the weight adjusters, respectively, a transmitter and a receiver connected to a first switch, a power distributor configured to connect the first switch to the respective signal transmitting and detecting units through internal wirings, and a compensator configured to the weights of the weight adjusters according to output signals of the signal transmitting and detecting units.

Abstract: A wireless transmitting and receiving device includes antennas, weight adjusters connected to antennas, respectively, and configured to apply weights to an amplitude and a phase of a signal communicating through the antennas, signal transmitting and detecting units connected to the weight adjusters, respectively, a transmitter and a receiver connected to a first switch, a power distributor configured to connect the first switch to the respective signal transmitting and detecting units through internal wirings, and a compensator configured to the weights of the weight adjusters according to output signals of the signal transmitting and detecting units.