Abstract: The neural network includes layers, and the layers each include a plurality of units, the plurality of units each have a weight coefficient associated with each input to a unit, and an importance parameter indicating importance of the weight coefficient. The method includes: in learning of a task, adjusting a first weight coefficient based on the importance parameter of the first weight coefficient and a first gradient of the first weight coefficient determined using a training set; and after the learning has been completed, determining, based on the training set, respective second gradients of a plurality of second weight coefficients included in a first layer including the first weight coefficient, and calculating the importance parameter of the first weight coefficient to be used in learning of a next task based on the respective second gradients.

Abstract: A point cloud coding device includes an interpolation unit configured to perform an interpolation process on a reconstructed point cloud of a coded frame with respect to attribute information in point cloud information and generating a reference frame of fractional precision, a motion estimation unit configured to perform motion estimation between the reference frame of fractional precision and the frame of integer precision to generate motion information, a prediction unit configured to generate a predicted value on the basis of the motion information, and an entropy coding unit configured to entropy-code the difference between a point cloud of the frame and the predicted value.

Abstract: Encoding efficiency for attribute information in point cloud information is improved. A point cloud encoding device includes a subsampling unit configured to preserve chroma signals of some points in a point cloud to be encoded and discard chroma signals of remaining points other than those points, and an attribute information encoding unit configured to encode the chroma signals of those points in the point cloud to be encoded and luma signals of all points.

Abstract: According to an aspect of the present disclosure, a method executed by an information processing apparatus in order to cause a neural network to learn a Tth task corresponding to a Tth learning set is provided. The method includes: for each of a plurality of units, determining an importance degree of the unit in the Tth task; for each of a plurality of layers, determining dissimilar tasks from among a first task to a (T?1)th task, the dissimilar tasks being not similar to the Tth task in terms of behaviors in the layer; and in learning that uses the Tth learning set, suppressing updating of weight parameters of the plurality of units included in the plurality of layers in accordance with importance degrees in the dissimilar tasks determined for each of the plurality of layers.

Abstract: The present embodiments relate to multi user detection in distributed antenna systems. A communication system may include a central processing circuitry and a plurality of access points connected to the central processing circuitry via a fronthaul interface. Each access point of the plurality of access points may include a plurality of antennas configured to receive signals from a plurality of UEs, and a local processing circuitry connected to the plurality of antennas. The local processing circuitry may be configured to estimate channel properties of a MIMO channel between the UEs and the antennas of the access point, and determine a subset of the UEs using the channel properties. The local processing circuitry may transmit data associated with the subset of wireless communication devices to the central processing circuitry for use to perform multiuser detection at the central processing circuitry.