4 results about "Minimum description length" patented technology

The present invention relates to a hybrid generative model device and an out-of-distribution determination method using same, wherein the hybrid generative model device comprises: a first input unit for inputting training data; a second input unit for inputting test data; a learning model unit for inputting the training data to a hybrid generative model to train the hybrid generative model, and inputting the test data to the hybrid generative model and allowing the hybrid generative model to produce an output; and an out-of-distribution (OOD) determination unit for determining OOD through a Wasserstein distance, which is measured according to the output of the hybrid generative model, between the training data and the test data and mutual information about the training data and the test data, and a minimal description length of the test data. Thereby, it is possible to effectively secure the integrity of an image as well as effectively determine in-distribution data and out-of-distribution data by using the hybrid generative model.

The invention discloses a destination reasoning method for urban vehicle tracking. The method comprises the following steps: firstly, carrying out adaptive segmentation on an original track by adopting an improved minimum description length algorithm to obtain a globally optimal track segment; then, through grid coding and a Base2vec model, the trajectory sequence is mapped into a low-dimensional distributed vector containing a geographic topological relation, namely a trajectory embedding sequence, and the sparse problem in spatial expression is relieved; then wavelet transform is applied to the track embedding sequence, and detailed information such as rapid turning or acceleration and deceleration in target movement is captured; and finally, an LSTM network is adopted, and a self-attention mechanism is integrated in the LSTM, so that the model can dynamically weight the output of different time steps according to the current trajectory segment, the network adaptively pays attention to key points in the trajectory, and the modeling capability for long-term dependence is enhanced. And finally, classifying the output of the LSTM network through a SoftMax classifier, converting a classification result into a specific geographic grid coordinate, and obtaining a predicted target destination position.

This invention provides a method for early warning of borehole wall collapse risk in impact-drilled pile foundations, belonging to the field of construction risk early warning technology. This invention collects multi-source signals by deploying an ultrasonic transducer array and a mechanical vibration acceleration sensor array at multiple depth nodes within the borehole. It extracts borehole wall collapse characteristic components from strong vibration interference using empirical mode decomposition and independent component analysis. Simultaneously, it inputs the multi-dimensional time-series data of the entire borehole into a minimum description length change point detection algorithm to detect abrupt changes in borehole wall state and outputs a hazard level score. Sensor data from each depth node is input into a physically embedded spatiotemporal graph attention model to output the collapse risk probability and estimated collapse depth location. Furthermore, an adaptive learning rate adjustment function driven by a stratum change intensity index dynamically adjusts the online update learning rate of the artificial intelligence model. Finally, it integrates multi-level early warning signals to output a final early warning conclusion, solving the technical problem of the inability to provide real-time and accurate early warning of borehole wall collapse risk during impact-drilled pile construction.

The application discloses a network representation method and system based on a multi-dimensional heterogeneous resource abstract view, and belongs to the technical field of communication networks, and comprises the following steps: acquiring information of network topology, equipment and link state, constructing a global original view, and initializing each node and link as a supernode and a superedge respectively to form an initial abstract view; generating a candidate supernode pair, calculating a comprehensive income of merging which takes into account storage compression income and path quality distortion, and iteratively performing supernode merging under the attribute purity constraint condition until the size of the abstract view meets a preset target; in the merging process, multi-dimensional attributes of the superedge are aggregated and updated, and only the superedge which reduces the overall storage overhead is reserved based on the minimum description length principle. While reducing the size of the network state, the application can maintain effective representation of multi-dimensional resource characteristics, has good dynamic adaptation capability, and is suitable for efficient modeling and analysis in a large-scale complex network environment.