arXiv subjects

Mingsheng Long

Publications and source records attributed to Mingsheng Long.

At least 127 records · Page 7Linked to original sources

arXiv2019

Learning to Transfer Examples for Partial Domain Adaptation

Domain adaptation is critical for learning in new and unseen environments. With domain adversarial training, deep networks can learn disentangled and transferable features that effectively diminish the dataset shift between the source and target domains for knowledge transfer. In the era of Big Data, the ready availability of large-scale labeled datasets has stimulated wide interest in partial domain adaptation (PDA), which transfers a recognizer from a labeled large domain to an unlabeled small domain. It extends standard domain adaptation to the scenario where target labels are only a subset of source labels. Under the condition that target labels are unknown, the key challenge of PDA is how to transfer relevant examples in the shared classes to promote positive transfer, and ignore irrelevant ones in the specific classes to mitigate negative transfer. In this work, we propose a unified approach to PDA, Example Transfer Network (ETN), which jointly learns domain-invariant representations across the source and target domains, and a progressive weighting scheme that quantifies the transferability of source examples while controlling their importance to the learning task in the target domain. A thorough evaluation on several benchmark datasets shows that our approach achieves state-of-the-art results for partial domain adaptation tasks.

cs.CV↗

arXiv2019

Spatiotemporal Pyramid Network for Video Action Recognition

Two-stream convolutional networks have shown strong performance in video action recognition tasks. The key idea is to learn spatiotemporal features by fusing convolutional networks spatially and temporally. However, it remains unclear how to model the correlations between the spatial and temporal structures at multiple abstraction levels. First, the spatial stream tends to fail if two videos share similar backgrounds. Second, the temporal stream may be fooled if two actions resemble in short snippets, though appear to be distinct in the long term. We propose a novel spatiotemporal pyramid network to fuse the spatial and temporal features in a pyramid structure such that they can reinforce each other. From the architecture perspective, our network constitutes hierarchical fusion strategies which can be trained as a whole using a unified spatiotemporal loss. A series of ablation experiments support the importance of each fusion strategy. From the technical perspective, we introduce the spatiotemporal compact bilinear operator into video analysis tasks. This operator enables efficient training of bilinear fusion operations which can capture full interactions between the spatial and temporal features. Our final network achieves state-of-the-art results on standard video datasets.

cs.CV↗

arXiv2019

Palladium Diselenide Long-Wavelength Infrared Photodetector with High Sensitivity and Stability

A long-wavelength infrared (IR) photodetector based on two-dimensional materials working at room temperature would have wide applications in many aspects in remote sensing, thermal imaging, biomedical optics, and medical imaging. However, sub-bandgap light detection in graphene and black phosphorus has been a long-standing scientific challenge because of low photoresponsivity, instability in the air and highdark current. In this study, we report a highly sensitive, air-stable and operable long-wavelength infrared photodetector at room temperature based on PdSe2 phototransistors and its heterostructure. A high photoresponsivity of ~42.1 AW-1 (at 10.6 μm) was demonstrated, which is an order of magnitude higher than the current record of platinum diselenide. Moreover, the dark current and noise power density were suppressed effectively by fabricating a van der Waals heterostructure. This work fundamentally contributes to establishing long-wavelength infrared detection by PdSe2 at the forefront of long-IR two-dimensional-materials-based photonics.

physics.app-ph↗

arXiv2019

Deep Triplet Quantization

Deep hashing establishes efficient and effective image retrieval by end-to-end learning of deep representations and hash codes from similarity data. We present a compact coding solution, focusing on deep learning to quantization approach that has shown superior performance over hashing solutions for similarity retrieval. We propose Deep Triplet Quantization (DTQ), a novel approach to learning deep quantization models from the similarity triplets. To enable more effective triplet training, we design a new triplet selection approach, Group Hard, that randomly selects hard triplets in each image group. To generate compact binary codes, we further apply a triplet quantization with weak orthogonality during triplet training. The quantization loss reduces the codebook redundancy and enhances the quantizability of deep representations through back-propagation. Extensive experiments demonstrate that DTQ can generate high-quality and compact binary codes, which yields state-of-the-art image retrieval performance on three benchmark datasets, NUS-WIDE, CIFAR-10, and MS-COCO.

cs.CV↗

arXiv2018

Conditional Adversarial Domain Adaptation

Adversarial learning has been embedded into deep networks to learn disentangled and transferable representations for domain adaptation. Existing adversarial domain adaptation methods may not effectively align different domains of multimodal distributions native in classification problems. In this paper, we present conditional adversarial domain adaptation, a principled framework that conditions the adversarial adaptation models on discriminative information conveyed in the classifier predictions. Conditional domain adversarial networks (CDANs) are designed with two novel conditioning strategies: multilinear conditioning that captures the cross-covariance between feature representations and classifier predictions to improve the discriminability, and entropy conditioning that controls the uncertainty of classifier predictions to guarantee the transferability. With theoretical guarantees and a few lines of codes, the approach has exceeded state-of-the-art results on five datasets.

cs.LG↗

arXiv2018

PredRNN++: Towards A Resolution of the Deep-in-Time Dilemma in Spatiotemporal Predictive Learning

We present PredRNN++, an improved recurrent network for video predictive learning. In pursuit of a greater spatiotemporal modeling capability, our approach increases the transition depth between adjacent states by leveraging a novel recurrent unit, which is named Causal LSTM for re-organizing the spatial and temporal memories in a cascaded mechanism. However, there is still a dilemma in video predictive learning: increasingly deep-in-time models have been designed for capturing complex variations, while introducing more difficulties in the gradient back-propagation. To alleviate this undesirable effect, we propose a Gradient Highway architecture, which provides alternative shorter routes for gradient flows from outputs back to long-range inputs. This architecture works seamlessly with causal LSTMs, enabling PredRNN++ to capture short-term and long-term dependencies adaptively. We assess our model on both synthetic and real video datasets, showing its ability to ease the vanishing gradient problem and yield state-of-the-art prediction results even in a difficult objects occlusion scenario.

cs.LG↗

arXiv2018

Deep Priority Hashing

Deep hashing enables image retrieval by end-to-end learning of deep representations and hash codes from training data with pairwise similarity information. Subject to the distribution skewness underlying the similarity information, most existing deep hashing methods may underperform for imbalanced data due to misspecified loss functions. This paper presents Deep Priority Hashing (DPH), an end-to-end architecture that generates compact and balanced hash codes in a Bayesian learning framework. The main idea is to reshape the standard cross-entropy loss for similarity-preserving learning such that it down-weighs the loss associated to highly-confident pairs. This idea leads to a novel priority cross-entropy loss, which prioritizes the training on uncertain pairs over confident pairs. Also, we propose another priority quantization loss, which prioritizes hard-to-quantize examples for generation of nearly lossless hash codes. Extensive experiments demonstrate that DPH can generate high-quality hash codes and yield state-of-the-art image retrieval results on three datasets, ImageNet, NUS-WIDE, and MS-COCO.

cs.CV↗

arXiv2018

Multi-Adversarial Domain Adaptation

Recent advances in deep domain adaptation reveal that adversarial learning can be embedded into deep networks to learn transferable features that reduce distribution discrepancy between the source and target domains. Existing domain adversarial adaptation methods based on single domain discriminator only align the source and target data distributions without exploiting the complex multimode structures. In this paper, we present a multi-adversarial domain adaptation (MADA) approach, which captures multimode structures to enable fine-grained alignment of different data distributions based on multiple domain discriminators. The adaptation can be achieved by stochastic gradient descent with the gradients computed by back-propagation in linear-time. Empirical evidence demonstrates that the proposed model outperforms state of the art methods on standard domain adaptation datasets.

cs.CV↗

arXiv2018

Negative Photoconductance in van der Waals Heterostructures-Based Floating Gate Phototransistor

Van der Waals (vdW) heterostructures made of two-dimensional materials have been demonstrated to be versatile architectures for optoelectronic applications due to strong light-matter interactions. However, most of light-controlled phenomena and applications in the vdW heterostructures rely on positive photoconductance (PPC). Negative photoconductance (NPC) has not yet been reported in vdW heterostructures. Here we report the observation of the NPC in ReS2/h-BN/MoS2 vdW heterostructures-based floating gate phototransistor. The fabricated devices exhibit excellent performance of nonvolatile memory without light illumination. More interestingly, we observe a gate-tunable transition between the PPC and the NPC under the light illumination. The observed NPC phenomenon can be attributed to the charge transfer between floating gate and conduction channel. Furthermore, we show that the control of NPC through light intensity is promising in realization of light-tunable multi-bit memory devices. Our results may enable potential applications in multifunctional memories and optoelectronic devices.

cond-mat.mes-hall↗

arXiv2018

Partial Adversarial Domain Adaptation

Domain adversarial learning aligns the feature distributions across the source and target domains in a two-player minimax game. Existing domain adversarial networks generally assume identical label space across different domains. In the presence of big data, there is strong motivation of transferring deep models from existing big domains to unknown small domains. This paper introduces partial domain adaptation as a new domain adaptation scenario, which relaxes the fully shared label space assumption to that the source label space subsumes the target label space. Previous methods typically match the whole source domain to the target domain, which are vulnerable to negative transfer for the partial domain adaptation problem due to the large mismatch between label spaces. We present Partial Adversarial Domain Adaptation (PADA), which simultaneously alleviates negative transfer by down-weighing the data of outlier source classes for training both source classifier and domain adversary, and promotes positive transfer by matching the feature distributions in the shared label space. Experiments show that PADA exceeds state-of-the-art results for partial domain adaptation tasks on several datasets.

cs.CV↗

arXiv2018

Analog Circuit Applications based on Ambipolar Graphene/MoTe2 Vertical Transistors

The current integrated circuit (IC) technology based on conventional MOS-FET (metal-oxide-semiconductor field-effect transistor) is approaching the limit of miniaturization with increasing demand on energy. Several analog circuit applications based on graphene FETs have been demonstrated with less components comparing to the conventional technology. However, low on/off current ratio caused by the semimetal nature of graphene has severely hindered its practical applications. Here we report a graphene/MoTe2 van der Waals (vdW) vertical transistor with V-shaped ambipolar field effect transfer characteristics to overcome this challenge. Investigations on temperature dependence of transport properties reveal that gate tunable asymmetric barriers of the devices are account for the ambipolar behaviors. Furthermore, to demonstrate the analog circuit applications of such vdW vertical transistors, we successfully realized output polarity controllable (OPC) amplifier and frequency doubler. These results enable vdW heterojunction based electronic devices to open up new possibilities for wide perspective in telecommunication field.

cond-mat.mes-hall↗

arXiv2017

Transfer Adversarial Hashing for Hamming Space Retrieval

Hashing is widely applied to large-scale image retrieval due to the storage and retrieval efficiency. Existing work on deep hashing assumes that the database in the target domain is identically distributed with the training set in the source domain. This paper relaxes this assumption to a transfer retrieval setting, which allows the database and the training set to come from different but relevant domains. However, the transfer retrieval setting will introduce two technical difficulties: first, the hash model trained on the source domain cannot work well on the target domain due to the large distribution gap; second, the domain gap makes it difficult to concentrate the database points to be within a small Hamming ball. As a consequence, transfer retrieval performance within Hamming Radius 2 degrades significantly in existing hashing methods. This paper presents Transfer Adversarial Hashing (TAH), a new hybrid deep architecture that incorporates a pairwise $t$-distribution cross-entropy loss to learn concentrated hash codes and an adversarial network to align the data distributions between the source and target domains. TAH can generate compact transfer hash codes for efficient image retrieval on both source and target domains. Comprehensive experiments validate that TAH yields state of the art Hamming space retrieval performance on standard datasets.

cs.CV↗

arXiv2017

Learning Multiple Tasks with Multilinear Relationship Networks

Deep networks trained on large-scale data can learn transferable features to promote learning multiple tasks. Since deep features eventually transition from general to specific along deep networks, a fundamental problem of multi-task learning is how to exploit the task relatedness underlying parameter tensors and improve feature transferability in the multiple task-specific layers. This paper presents Multilinear Relationship Networks (MRN) that discover the task relationships based on novel tensor normal priors over parameter tensors of multiple task-specific layers in deep convolutional networks. By jointly learning transferable features and multilinear relationships of tasks and features, MRN is able to alleviate the dilemma of negative-transfer in the feature layers and under-transfer in the classifier layer. Experiments show that MRN yields state-of-the-art results on three multi-task learning datasets.

cs.LG↗

arXiv2017

Deep Transfer Learning with Joint Adaptation Networks

Deep networks have been successfully applied to learn transferable features for adapting models from a source domain to a different target domain. In this paper, we present joint adaptation networks (JAN), which learn a transfer network by aligning the joint distributions of multiple domain-specific layers across domains based on a joint maximum mean discrepancy (JMMD) criterion. Adversarial training strategy is adopted to maximize JMMD such that the distributions of the source and target domains are made more distinguishable. Learning can be performed by stochastic gradient descent with the gradients computed by back-propagation in linear-time. Experiments testify that our model yields state of the art results on standard datasets.

cs.LG↗

arXiv2017

HashNet: Deep Learning to Hash by Continuation

Learning to hash has been widely applied to approximate nearest neighbor search for large-scale multimedia retrieval, due to its computation efficiency and retrieval quality. Deep learning to hash, which improves retrieval quality by end-to-end representation learning and hash encoding, has received increasing attention recently. Subject to the ill-posed gradient difficulty in the optimization with sign activations, existing deep learning to hash methods need to first learn continuous representations and then generate binary hash codes in a separated binarization step, which suffer from substantial loss of retrieval quality. This work presents HashNet, a novel deep architecture for deep learning to hash by continuation method with convergence guarantees, which learns exactly binary hash codes from imbalanced similarity data. The key idea is to attack the ill-posed gradient problem in optimizing deep networks with non-smooth binary activations by continuation method, in which we begin from learning an easier network with smoothed activation function and let it evolve during the training, until it eventually goes back to being the original, difficult to optimize, deep network with the sign activation function. Comprehensive empirical evidence shows that HashNet can generate exactly binary hash codes and yield state-of-the-art multimedia retrieval performance on standard benchmarks.

cs.LG↗

arXiv2017

Partial Transfer Learning with Selective Adversarial Networks

Adversarial learning has been successfully embedded into deep networks to learn transferable features, which reduce distribution discrepancy between the source and target domains. Existing domain adversarial networks assume fully shared label space across domains. In the presence of big data, there is strong motivation of transferring both classification and representation models from existing big domains to unknown small domains. This paper introduces partial transfer learning, which relaxes the shared label space assumption to that the target label space is only a subspace of the source label space. Previous methods typically match the whole source domain to the target domain, which are prone to negative transfer for the partial transfer problem. We present Selective Adversarial Network (SAN), which simultaneously circumvents negative transfer by selecting out the outlier source classes and promotes positive transfer by maximally matching the data distributions in the shared label space. Experiments demonstrate that our models exceed state-of-the-art results for partial transfer learning tasks on several benchmark datasets.

cs.LG↗

arXiv2017

Intrinsic p-type W-based transition metal dichalcogenide by substitutional Ta-doping

Two-dimensional (2D) transition metal dichalcogenides (TMDs) have recently emerged as promising candidates for future electronics and optoelectronics. While most of TMDs are intrinsic n-type semiconductors due to electron donating which originates from chalcogen vacancies, obtaining intrinsic high-quality p-type semiconducting TMDs has been challenging. Here, we report an experimental approach to obtain intrinsic p-type Tungsten (W)-based TMDs by substitutional Ta-doping. The obtained few-layer Ta-doped WSe2 (Ta0.01W0.99Se2) field-effect transistor (FET) devices exhibit competitive p-type performances, including ~10^6 current on/off at room temperature. We also demonstrate high quality van der Waals (vdW) p-n heterojunctions based on Ta0.01W0.99Se2/MoS2 structure, which exhibit nearly ideal diode characteristics (with an ideality factor approaching 1 and a rectification ratio up to 10^5) and excellent photodetecting performance. Our study suggests that substitutional Ta-doping holds great promise to realize intrinsic p-type W-based TMDs for future electronic and photonic applications.

cond-mat.mtrl-sci↗

arXiv2017

Room-temperature high detectivity mid-infrared photodetectors based on black arsenic phosphorus

The mid-infrared (MIR) spectral range, pertaining to important applications such as molecular 'fingerprint' imaging, remote sensing, free space telecommunication and optical radar, is of particular scientific interest and technological importance. However, state-of-the-art materials for MIR detection are limited by intrinsic noise and inconvenient fabrication processes, resulting in high cost photodetectors requiring cryogenic operation. We report black arsenic-phosphorus-based long wavelength infrared photodetectors with room temperature operation up to 8.2 um, entering the second MIR atmospheric transmission window. Combined with a van der Waals heterojunction, room temperature specific detectivity higher than 4.9*10^9 Jones was obtained in the 3-5 um range. The photodetector works in a zero-bias photovoltaic mode, enabling fast photoresponse and low dark noise. Our van der Waals heterojunction photodector not only exemplify black arsenic-phosphorus as a promising candidate for MIR opto-electronic applications, but also pave the way for a general strategy to suppress 1/f noise in photonic devices.

cond-mat.mtrl-sci↗