Purpose: Although recent deep energy-based generative models (EBMs) have shown encouraging results in many image generation tasks, how to take advantage of the self-adversarial cogitation in deep EBMs to boost the performance of Magnetic Resonance Im
aging (MRI) reconstruction is still desired. Methods: With the successful application of deep learning in a wide range of MRI reconstruction, a line of emerging research involves formulating an optimization-based reconstruction method in the space of a generative model. Leveraging this, a novel regularization strategy is introduced in this article which takes advantage of self-adversarial cogitation of the deep energy-based model. More precisely, we advocate for alternative learning a more powerful energy-based model with maximum likelihood estimation to obtain the deep energy-based information, represented as image prior. Simultaneously, implicit inference with Langevin dynamics is a unique property of re-construction. In contrast to other generative models for reconstruction, the proposed method utilizes deep energy-based information as the image prior in reconstruction to improve the quality of image. Results: Experiment results that imply the proposed technique can obtain remarkable performance in terms of high reconstruction accuracy that is competitive with state-of-the-art methods, and does not suffer from mode collapse. Conclusion: Algorithmically, an iterative approach was presented to strengthen EBM training with the gradient of energy network. The robustness and the reproducibility of the algorithm were also experimentally validated. More importantly, the proposed reconstruction framework can be generalized for most MRI reconstruction scenarios.
In the present article, we use Robbas method to give an estimation of the Newton polygon for the L function on torus.
As an effective way to integrate the information contained in multiple medical images under different modalities, medical image synthesis and fusion have emerged in various clinical applications such as disease diagnosis and treatment planning. In th
is paper, an invertible and variable augmented network (iVAN) is proposed for medical image synthesis and fusion. In iVAN, the channel number of the network input and output is the same through variable augmentation technology, and data relevance is enhanced, which is conducive to the generation of characterization information. Meanwhile, the invertible network is used to achieve the bidirectional inference processes. Due to the invertible and variable augmentation schemes, iVAN can not only be applied to the mappings of multi-input to one-output and multi-input to multi-output, but also be applied to one-input to multi-output. Experimental results demonstrated that the proposed method can obtain competitive or superior performance in comparison to representative medical image synthesis and fusion methods.
We analyze a game of technology development where players allocate resources between exploration, which continuously expands the public domain of available technologies, and exploitation, which yields a flow payoff by adopting the explored technologi
es. The qualities of the technologies are correlated and initially unknown, and this uncertainty is fully resolved once the technologies are explored. We consider Markov perfect equilibria with the quality difference between the best available technology and the latest technology under development as the state variable. In all such equilibria, while the players do not fully internalize the benefit of failure owing to free-riding incentives, they are more tolerant of failure than in the single-agent optimum thanks to an encouragement effect. In the unique symmetric equilibrium, the cost of exploration determines whether free-riding prevails as team size grows. Pareto improvements over the symmetric equilibrium can be achieved by asymmetric equilibria where players take turns performing exploration.
This paper aims to investigate the characteristics of durations of discretionary lane changes (LCs) on freeways based on an enriched dataset containing LC vehicle trajectories of 2905 passenger cars and 433 heavy vehicles. A comprehensive analysis of
LC duration is conducted and four stochastic LC duration models are built according to vehicle types and LC directions. It is found that the LC duration varies across different vehicle types and LC directions. The modelling results show that different variables have different effects on LC duration for different vehicle types and LC directions. Fixed-parameter, latent class, and random parameter accelerated hazard time (AFT) models were built considering driver heterogeneity. Results show that heavy vehicle drivers show more heterogeneity. Different variables were found for different vehicle types and LC directions. The results of this study can be beneficial to understand the mechanism of LC process and the influence of LC on traffic flow.
This work presents an unsupervised deep learning scheme that exploiting high-dimensional assisted score-based generative model for color image restoration tasks. Considering that the sample number and internal dimension in score-based generative mode
l have key influence on estimating the gradients of data distribution, two different high-dimensional ways are proposed: The channel-copy transformation increases the sample number and the pixel-scale transformation decreases feasible space dimension. Subsequently, a set of high-dimensional tensors represented by these transformations are used to train the network through denoising score matching. Then, sampling is performed by annealing Langevin dynamics and alternative data-consistency update. Furthermore, to alleviate the difficulty of learning high-dimensional representation, a progressive strategy is proposed to leverage the performance. The proposed unsupervised learning and iterative restoration algo-rithm, which involves a pre-trained generative network to obtain prior, has transparent and clear interpretation compared to other data-driven approaches. Experimental results on demosaicking and inpainting conveyed the remarkable performance and diversity of our proposed method.
Unsupervised deep learning has recently demonstrated the promise to produce high-quality samples. While it has tremendous potential to promote the image colorization task, the performance is limited owing to the manifold hypothesis in machine learnin
g. This study presents a novel scheme that exploiting the score-based generative model in wavelet domain to address the issue. By taking advantage of the multi-scale and multi-channel representation via wavelet transform, the proposed model learns the priors from stacked wavelet coefficient components, thus learns the image characteristics under coarse and detail frequency spectrums jointly and effectively. Moreover, such a highly flexible generative model without adversarial optimization can execute colorization tasks better under dual consistency terms in wavelet domain, namely data-consistency and structure-consistency. Specifically, in the training phase, a set of multi-channel tensors consisting of wavelet coefficients are used as the input to train the network by denoising score matching. In the test phase, samples are iteratively generated via annealed Langevin dynamics with data and structure consistencies. Experiments demonstrated remarkable improvements of the proposed model on colorization quality, particularly on colorization robustness and diversity.
Observationally, the interstellar gas-phase abundance of deuterium (D) is considerably depleted and the missing D atoms are often postulated to have been locked up into carbonaceous solids and polycyclic aromatic hydrocarbon (PAH) molecules. An accur
ate knowledge of the fractional amount of D (relative to H) tied up in carbon dust and PAHs has important cosmological implications since D originated exclusively from the Big Bang and the present-day D abundance, after accounting for the astration it has experienced during the Galactic evolution, provides essential clues to the primordial nucleosynthesis and the cosmological parameters. To quantitatively explore the extent to which PAHs could possibly accommodate the observed D depletion, we have previously quantum-chemically computed the infrared vibrational spectra of mono-deuterated PAHs and derived the mean intrinsic band strengths of the 3.3 $mu$m C--H stretch (A$_{3.3}$) and the 4.4 $mu$m C--D stretch (A$_{4.4}$). Here we extend our previous work to multi-deuterated PAH species of different deuterations, sizes and structures. We find that both the intrinsic band strengths A$_{3.3}$ and A$_{4.4}$ and their ratios A$_{4.4}$/A$_{3.3}$ not only show little variations among PAHs of different deuterations, sizes and structures, they are also closely similar to that of mono-deuterated PAHs. Therefore, a PAH deuteration level (i.e., the fraction of peripheral atoms attached to C atoms in the form of D) of ~2.4% previously estimated from the observed 4.4 $mu$m to 3.3 $mu$m band ratio based on the A$_{4.4}$/A$_{3.3}$ ratio of mono-deuterated PAHs is robust.
Prototype learning is extensively used for few-shot segmentation. Typically, a single prototype is obtained from the support feature by averaging the global object information. However, using one prototype to represent all the information may lead to
ambiguities. In this paper, we propose two novel modules, named superpixel-guided clustering (SGC) and guided prototype allocation (GPA), for multiple prototype extraction and allocation. Specifically, SGC is a parameter-free and training-free approach, which extracts more representative prototypes by aggregating similar feature vectors, while GPA is able to select matched prototypes to provide more accurate guidance. By integrating the SGC and GPA together, we propose the Adaptive Superpixel-guided Network (ASGNet), which is a lightweight model and adapts to object scale and shape variation. In addition, our network can easily generalize to k-shot segmentation with substantial improvement and no additional computational cost. In particular, our evaluations on COCO demonstrate that ASGNet surpasses the state-of-the-art method by 5% in 5-shot segmentation.
Needle-like metallic particles have been suggested to explain a wide variety of astrophysical phenomena, ranging from the mid-infrared interstellar extinction to the thermalization of starlight to generate the cosmic microwave background. These sugge
stions rely on the amplitude and the wavelength dependence of the absorption cross sections of metallic needles. On the absence of an exact solution to the absorption properties of metallic needles, their absorption cross sections are often derived from the antenna approximation. However, it is shown here that the antenna approximation is not an appropriate representation since it violates the Kramers-Kronig relation. Stimulated by the recent discovery of iron whiskers in asteroid Itokawa and graphite whiskers in carbonaceous chondrites, we call for rigorous calculations of the absorption cross sections of metallic needle-like particles, presumably with the discrete dipole approximation. We also call for experimental studies of the formation and growth mechanisms of metallic needle-like particles as well as experimental measurements of the absorption cross sections of metallic needles of various aspect ratios over a wide wavelength range to bound theoretical calculations.
