arxiv-topics-distilbert-base-cased

Try a prompt

• Title: ImageNet classification with deep convolutional neural networks Abstract: We trained a large, deep convolutional neural network to classify the 1.2 million high-resolution images in the ImageNet LSVRC-2010 contest into the 1000 different classes. On the test data, we achieved top-1 and top-5 error rates of 37.5% and 17.0% which is considerably better than the previous state-of-the-art. The neural network, which has 60 million parameters and 650,000 neurons, consists of five convolutional layers, some of which are followed by max-pooling layers, and three fully-connected layers with a final 1000-way softmax. To make training faster, we used non-saturating neurons and a very efficient GPU implementation of the convolution operation. To reduce overriding in the fully-connected layers we employed a recently-developed regularization method called "dropout" that proved to be very effective. We also entered a variant of this model in the ILSVRC-2012 competition and achieved a winning top-5 test error rate of 15.3%, compared to 26.2\% achieved by the second-best entry. Authors: A. Krizhevsky, Ilya Sutskever, Geoffrey E. HintonCopy
• Title: The Design of Radio Telescope Array Configurations using Multiobjective Optimization: Imaging Performance versus Cable Length Abstract: The next generation of radio telescope interferometric arrays requires careful design of the array configuration to optimize the performance of the overall system. We have developed a framework, based on a genetic algorithm, for rapid exploration and optimization of the objective space pertaining to multiple objectives. We have evaluated a large space of possible designs for 27-, 60-, 100-, and 160-station arrays. The 27-station optimizations can be compared to the well-known VLA case, and the larger array designs apply to arrays currently under design such as LOFAR, ATA, and the SKA. In the initial implementation of our framework we evaluate designs with respect to two metrics, array imaging performance and the length of cable necessary to connect the stations. Imaging performance is measured by the degree to which the sampling of the uv plane is uniform. For the larger arrays we find that well-known geometric designs perform well and occupy the Pareto front of optimum solutions. For the 27-element case we find designs, combining features of the well-known designs, that are more optimal as measured by these two metrics. The results obtained by the multiobjective genetic optimization are corroborated by simulated annealing, which also reveals the role of entropy in array optimization. Our framework is general, and may be applied to other design goals and issues, such as particular schemes for sampling the uv plane, array robustness, and phased deployment of arrays. Authors: Babak E. Cohanim, Jacqueline N. Hewitt, Olivier de WeckCopy
• Title: Evidence for a Neutron Star in the non-pulsating massive X-ray binary 4U2206+54 Abstract: We present an analysis of archival RXTE and BeppoSAX data of the X-ray source 4U2206+54 . For the first time, high energy data (> 30 kev) are analyzed for this source. The data are well described by comptonization models (CompTT and BMC) in which seed photons with temperatures between 1.1 kev and 1.5 kev are comptonized by a hot plasma at 50 kev thereby producing a hard tail which extends up to, at least, 100 kev. We offer a new method of identification of neutron star systems using a temperature - luminosity relation. If a given X-ray source is characterized by a low bolometric luminosity and a relatively high color blackbody temperature (>1 kev) it has necessarily to be a neutron star rather than a black hole. From these arguments it is shown that the area of the soft photon source must be small (r ~ 1 km) and that the accretion disk, if present, must be truncated very far from the compact object. Here we report on the possible existence of a cyclotron line around 30 kev. The presence of a neutron star in the system is strongly favored by the available data. Authors: J. M. Torrejón, I. Kreykenbohm, A. Orr, L. Titarchuk, I. NegueruelaCopy
• Title: Solving the Schrodinger Equation for a Quantum Well with a Non-Uniform Potential Abstract: We present a numerical solution to the Schrodinger equation for a quantum well with a non-uniform potential. The potential is a Gaussian with a non-uniform distribution of energies. The solution is a solution to the Schrodinger equation with a non-uniform potential. The solution is a non-uniform solution to the Schrodinger equation with a non-uniform potential. Authors: George K. Kostopoulos, John A. Kostopoulos, and John C. KostopoulosCopy