Light dark bosons in the JUNO-TAO neutrino detector

This work presents a sensitivity study of a reactor liquid scintillator detector to three kinds of dark bosons with masses below 1 MeV, such as dark photons, axion-like particles and light scalar bosons. The JUNO-TAO detector with Taishan nuclear reactor is taken as a reference. With proposed 180 days data taking, the sensitivity to the dark bosons can reach $sim10^{-5}$ 95%C.L. for the optimized signal to background ratio for the electron coupling constant $it{g_X} $ through inverse Compton-like scattering. The background systematic uncertainty presents as the main limiting factor for the further sensitivity improvement. Additionally the differential and the inverse differential cross sections have been derived for all three boson types and their interactions with electrons in liquid scintillator.

Knowledge Distillation Using Hierarchical Self-Supervision Augmented Distribution

Knowledge distillation (KD) is an effective framework that aims to transfer meaningful information from a large teacher to a smaller student. Generally, KD often involves how to define and transfer knowledge. Previous KD methods often focus on mining various forms of knowledge, for example, feature maps and refined information. However, the knowledge is derived from the primary supervised task and thus is highly task-specific. Motivated by the recent success of self-supervised representation learning, we propose an auxiliary self-supervision augmented task to guide networks to learn more meaningful features. Therefore, we can derive soft self-supervision augmented distributions as richer dark knowledge from this task for KD. Unlike previous knowledge, this distribution encodes joint knowledge from supervised and self-supervised feature learning. Beyond knowledge exploration, another crucial aspect is how to learn and distill our proposed knowledge effectively. To fully take advantage of hierarchical feature maps, we propose to append several auxiliary branches at various hidden layers. Each auxiliary branch is guided to learn self-supervision augmented task and distill this distribution from teacher to student. Thus we call our KD method as Hierarchical Self-Supervision Augmented Knowledge Distillation (HSSAKD). Experiments on standard image classification show that both offline and online HSSAKD achieves state-of-the-art performance in the field of KD. Further transfer experiments on object detection further verify that HSSAKD can guide the network to learn better features, which can be attributed to learn and distill an auxiliary self-supervision augmented task effectively.

Improving Speaker Identification for Shared Devices by Adapting Embeddings to Speaker Subsets

Speaker identification typically involves three stages. First, a front-end speaker embedding model is trained to embed utterance and speaker profiles. Second, a scoring function is applied between a runtime utterance and each speaker profile. Finally, the speaker is identified using nearest neighbor according to the scoring metric. To better distinguish speakers sharing a device within the same household, we propose a household-adapted nonlinear mapping to a low dimensional space to complement the global scoring metric. The combined scoring function is optimized on labeled or pseudo-labeled speaker utterances. With input dropout, the proposed scoring model reduces EER by 45-71% in simulated households with 2 to 7 hard-to-discriminate speakers per household. On real-world internal data, the EER reduction is 49.2%. From t-SNE visualization, we also show that clusters formed by household-adapted speaker embeddings are more compact and uniformly distributed, compared to clusters formed by global embeddings before adaptation.

Online Learning of Independent Cascade Models with Node-level Feedback

We propose a detailed analysis of the online-learning problem for Independent Cascade (IC) models under node-level feedback. These models have widespread applications in modern social networks. Existing works for IC models have only shed light on edge-level feedback models, where the agent knows the explicit outcome of every observed edge. Little is known about node-level feedback models, where only combined outcomes for sets of edges are observed; in other words, the realization of each edge is censored. This censored information, together with the nonlinear form of the aggregated influence probability, make both parameter estimation and algorithm design challenging. We establish the first confidence-region result under this setting. We also develop an online algorithm achieving a cumulative regret of $mathcal{O}( sqrt{T})$, matching the theoretical regret bound for IC models with edge-level feedback.

Do current X-ray observations capture most of the black-hole accretion at high redshifts?

The cosmic black hole accretion density (BHAD) is critical for our understanding of the formation and evolution of supermassive black holes (BHs). However, at high redshifts ($z>3$), X-ray observations report BHADs significantly ($sim 10$ times) lower than those predicted by cosmological simulations. It is therefore paramount to constrain the high-$z$ BHAD using independent methods other than direct X-ray detections. The recently established relation between star formation rate and BH accretion rate among bulge-dominated galaxies provides such a chance, as it enables an estimate of the BHAD from the star-formation histories (SFHs) of lower-redshift objects. Using the CANDELS Lyman-$alpha$ Emission At Reionization (CLEAR) survey, we model the SFHs for a sample of 108 bulge-dominated galaxies at $z=$0.7-1.5, and further estimate the BHAD contributed by their high-$z$ progenitors. The predicted BHAD at $zapprox 4$-5 is consistent with the simulation-predicted values, but higher than the X-ray measurements (by $approx$3-10 times at $z=$4-5). Our result suggests that the current X-ray surveys could be missing many heavily obscured Compton-thick active galactic nuclei (AGNs) at high redshifts. However, this BHAD estimation assumes that the high-$z$ progenitors of our $z=$0.7-1.5 sample remain bulge-dominated where star formation is correlated with BH cold-gas accretion. Alternatively, our prediction could signify a stark decline in the fraction of bulges in high-$z$ galaxies (with an associated drop in BH accretion). JWST and Origins will resolve the discrepancy between our predicted BHAD and the X-ray results by constraining Compton-thick AGN and bulge evolution at high redshifts.

Unknown Object Segmentation through Domain Adaptation

The ability to segment unknown objects in cluttered scenes has a profound impact on robot grasping. The rise of deep learning has greatly transformed the pipeline of robotic grasping from model-based approach to data-driven stream, which generally requires a large scale of grasping data either collected in simulation or from real-world examples. In this paper, we proposed a sim-to-real framework to transfer the object segmentation model learned in simulation to the real-world. First, data samples are collected in simulation, including RGB, 6D pose, and point cloud. Second, we also present a GAN-based unknown object segmentation method through domain adaptation, which consists of an image translation module and an image segmentation module. The image translation module is used to shorten the reality gap and the segmentation module is responsible for the segmentation mask generation. We used the above method to perform segmentation experiments on unknown objects in a bin-picking scenario. Finally, the experimental result shows that the segmentation model learned in simulation can be used for real-world data segmentation.

Hierarchical Self-supervised Augmented Knowledge Distillation

Knowledge distillation often involves how to define and transfer knowledge from teacher to student effectively. Although recent self-supervised contrastive knowledge achieves the best performance, forcing the network to learn such knowledge may damage the representation learning of the original class recognition task. We therefore adopt an alternative self-supervised augmented task to guide the network to learn the joint distribution of the original recognition task and self-supervised auxiliary task. It is demonstrated as a richer knowledge to improve the representation power without losing the normal classification capability. Moreover, it is incomplete that previous methods only transfer the probabilistic knowledge between the final layers. We propose to append several auxiliary classifiers to hierarchical intermediate feature maps to generate diverse self-supervised knowledge and perform the one-to-one transfer to teach the student network thoroughly. Our method significantly surpasses the previous SOTA SSKD with an average improvement of 2.56% on CIFAR-100 and an improvement of 0.77% on ImageNet across widely used network pairs. Codes are available at https://github.com/winycg/HSAKD.

3D AGSE-VNet: An Automatic Brain Tumor MRI Data Segmentation Framework

Background: Glioma is the most common brain malignant tumor, with a high morbidity rate and a mortality rate of more than three percent, which seriously endangers human health. The main method of acquiring brain tumors in the clinic is MRI. Segmentation of brain tumor regions from multi-modal MRI scan images is helpful for treatment inspection, post-diagnosis monitoring, and effect evaluation of patients. However, the common operation in clinical brain tumor segmentation is still manual segmentation, lead to its time-consuming and large performance difference between different operators, a consistent and accurate automatic segmentation method is urgently needed. Methods: To meet the above challenges, we propose an automatic brain tumor MRI data segmentation framework which is called AGSE-VNet. In our study, the Squeeze and Excite (SE) module is added to each encoder, the Attention Guide Filter (AG) module is added to each decoder, using the channel relationship to automatically enhance the useful information in the channel to suppress the useless information, and use the attention mechanism to guide the edge information and remove the influence of irrelevant information such as noise. Results: We used the BraTS2020 challenge online verification tool to evaluate our approach. The focus of verification is that the Dice scores of the whole tumor (WT), tumor core (TC) and enhanced tumor (ET) are 0.68, 0.85 and 0.70, respectively. Conclusion: Although MRI images have different intensities, AGSE-VNet is not affected by the size of the tumor, and can more accurately extract the features of the three regions, it has achieved impressive results and made outstanding contributions to the clinical diagnosis and treatment of brain tumor patients.

A double-pivot degenerate-tolerable simplex algorithm for linear programming

A double pivot algorithm that combines features of two recently published papers by these authors is proposed. The proposed algorithm is implemented in MATLAB. The MATLAB code is tested, along with a MATLAB implementation of Dantzigs algorithm, for several test sets, including a set of cycling LP problems, Klee-Mintys problems, randomly generated linear programming (LP) problems, and Netlib benchmark problems. The test result shows that the proposed algorithm is (a) degenerate-tolerance as we expected, and (b) more efficient than Dantzigs algorithm for large size randomly generated LP problems but less efficient for Netlib benchmark problems and small size randomly generated problems in terms of CPU time.

RingFed: Reducing Communication Costs in Federated Learning on Non-IID Data

Federated learning is a widely used distributed deep learning framework that protects the privacy of each client by exchanging model parameters rather than raw data. However, federated learning suffers from high communication costs, as a considerable number of model parameters need to be transmitted many times during the training process, making the approach inefficient, especially when the communication network bandwidth is limited. This article proposes RingFed, a novel framework to reduce communication overhead during the training process of federated learning. Rather than transmitting parameters between the center server and each client, as in original federated learning, in the proposed RingFed, the updated parameters are transmitted between each client in turn, and only the final result is transmitted to the central server, thereby reducing the communication overhead substantially. After several local updates, clients first send their parameters to another proximal client, not to the center server directly, to preaggregate. Experiments on two different public datasets show that RingFed has fast convergence, high model accuracy, and low communication cost.