Search

Deep learning has revolutionized artificial intelligence (AI), achieving remarkable progress in fields such as computer vision, speech recognition, and natural language processing. Moreover, the recent success of large language models (LLMs) has fueled a surge in research on large-scale neural networks. However, the escalating demand for computing resources and energy consumption has prompted the search for energy-efficient alternatives. Inspired by the human brain, spiking neural networks (SNNs) promise energy-efficient computation with event-driven spikes. To provide future directions toward building energy-efficient large SNN models, we present a survey of existing methods for developing deep spiking neural networks, with a focus on emerging Spiking Transformers. Our main contributions are as follows: (1) an overview of learning methods for deep spiking neural networks, categorized by ANN-to-SNN conversion and direct training with surrogate gradients; (2) an overview of network architectures for deep spiking neural networks, categorized by deep convolutional neural networks (DCNNs) and Transformer architecture; and (3) a comprehensive comparison of state-of-the-art deep SNNs with a focus on emerging Spiking Transformers. We then further discuss and outline future directions toward large-scale SNNs.

Although recent advancements in text-to-3D generation have significantly improved generation quality, issues like limited level of detail and low fidelity still persist, which requires further improvement. To understand the essence of those issues, we thoroughly analyze current score distillation methods by connecting theories of consistency distillation to score distillation. Based on the insights acquired through analysis, we propose an optimization framework, Guided Consistency Sampling (GCS), integrated with 3D Gaussian Splatting (3DGS) to alleviate those issues. Additionally, we have observed the persistent oversaturation in the rendered views of generated 3D assets. From experiments, we find that it is caused by unwanted accumulated brightness in 3DGS during optimization. To mitigate this issue, we introduce a Brightness-Equalized Generation (BEG) scheme in 3DGS rendering. Experimental results demonstrate that our approach generates 3D assets with more details and higher fidelity than state-of-the-art methods. The codes are released at https://github.com/LMozart/ECCV2024-GCS-BEG., Comment: Paper accepted by ECCV2024

Policy-based methods have achieved remarkable success in solving challenging reinforcement learning problems. Among these methods, off-policy policy gradient methods are particularly important due to that they can benefit from off-policy data. However, these methods suffer from the high variance of the off-policy policy gradient (OPPG) estimator, which results in poor sample efficiency during training. In this paper, we propose an off-policy policy gradient method with the optimal action-dependent baseline (Off-OAB) to mitigate this variance issue. Specifically, this baseline maintains the OPPG estimator's unbiasedness while theoretically minimizing its variance. To enhance practical computational efficiency, we design an approximated version of this optimal baseline. Utilizing this approximation, our method (Off-OAB) aims to decrease the OPPG estimator's variance during policy optimization. We evaluate the proposed Off-OAB method on six representative tasks from OpenAI Gym and MuJoCo, where it demonstrably surpasses state-of-the-art methods on the majority of these tasks., Comment: 12 pages, 3 figures

Natural Light Uncalibrated Photometric Stereo (NaUPS) relieves the strict environment and light assumptions in classical Uncalibrated Photometric Stereo (UPS) methods. However, due to the intrinsic ill-posedness and high-dimensional ambiguities, addressing NaUPS is still an open question. Existing works impose strong assumptions on the environment lights and objects' material, restricting the effectiveness in more general scenarios. Alternatively, some methods leverage supervised learning with intricate models while lacking interpretability, resulting in a biased estimation. In this work, we proposed Spin Light Uncalibrated Photometric Stereo (Spin-UP), an unsupervised method to tackle NaUPS in various environment lights and objects. The proposed method uses a novel setup that captures the object's images on a rotatable platform, which mitigates NaUPS's ill-posedness by reducing unknowns and provides reliable priors to alleviate NaUPS's ambiguities. Leveraging neural inverse rendering and the proposed training strategies, Spin-UP recovers surface normals, environment light, and isotropic reflectance under complex natural light with low computational cost. Experiments have shown that Spin-UP outperforms other supervised / unsupervised NaUPS methods and achieves state-of-the-art performance on synthetic and real-world datasets. Codes and data are available at https://github.com/LMozart/CVPR2024-SpinUP., Comment: Paper accepted by CVPR2024

Foreground removal is one of the biggest challenges in the detection of the Cosmic Dawn (CD) and Epoch of Reionization (EoR). Various foreground subtraction techniques have been developed based on the spectral smoothness of foregrounds. However, the sources with a spectral peak (SP) at Megahertz may break down the spectral smoothness at low frequencies (< 1000 MHz). In this paper, we cross-match the GaLactic and Extragalactic All-sky Murchison Widefield Array (GLEAM) extragalactic source catalogue with three other radio source catalogues, covering the frequency range from 72 MHz to 1.4 GHz, to search for sources with spectral turnover. 4,423 sources from the GLEAM catalogue are identified as SP sources, representing approximately 3.2 per cent of the GLEAM radio source population. We utilize the properties of SP source candidates obtained from real observations to establish simulations and test the impact of SP sources on the extraction of CD/EoR signals. We statistically compare the differences introduced by SP sources in the residuals after removing the foregrounds with three methods, which are polynomial fitting, Principal Component Analysis (PCA), and fast independent component analysis (FastICA). Our results indicate that the presence of SP sources in the foregrounds has a negligible influence on extracting the CD/EoR signal. After foreground subtraction, the contribution from SP sources to the total power in the two-dimensional (2D) power spectrum within the EoR window is approximately 3 to 4 orders of magnitude lower than the CD/EoR signal., Comment: 14 pages, 14 figures

Recent advancement in computer vision has significantly lowered the barriers to artistic creation. Exemplar-based image translation methods have attracted much attention due to flexibility and controllability. However, these methods hold assumptions regarding semantics or require semantic information as the input, while accurate semantics is not easy to obtain in artistic images. Besides, these methods suffer from cross-domain artifacts due to training data prior and generate imprecise structure due to feature compression in the spatial domain. In this paper, we propose an arbitrary Style Image Manipulation Network (SIM-Net), which leverages semantic-free information as guidance and a region transportation strategy in a self-supervised manner for image generation. Our method balances computational efficiency and high resolution to a certain extent. Moreover, our method facilitates zero-shot style image manipulation. Both qualitative and quantitative experiments demonstrate the superiority of our method over state-of-the-art methods.Code is available at https://github.com/SnailForce/SIM-Net.

Panoramic imaging research on geometry recovery and High Dynamic Range (HDR) reconstruction becomes a trend with the development of Extended Reality (XR). Neural Radiance Fields (NeRF) provide a promising scene representation for both tasks without requiring extensive prior data. However, in the case of inputting sparse Low Dynamic Range (LDR) panoramic images, NeRF often degrades with under-constrained geometry and is unable to reconstruct HDR radiance from LDR inputs. We observe that the radiance from each pixel in panoramic images can be modeled as both a signal to convey scene lighting information and a light source to illuminate other pixels. Hence, we propose the irradiance fields from sparse LDR panoramic images, which increases the observation counts for faithful geometry recovery and leverages the irradiance-radiance attenuation for HDR reconstruction. Extensive experiments demonstrate that the irradiance fields outperform state-of-the-art methods on both geometry recovery and HDR reconstruction and validate their effectiveness. Furthermore, we show a promising byproduct of spatially-varying lighting estimation. The code is available at https://github.com/Lu-Zhan/Pano-NeRF.

Detecting the cosmic 21 cm signal from Epoch of Reionization (EoR) has always been a difficult task. Although the Galactic foreground can be regarded as a smooth power-law spectrum, due to the chromaticity of the antenna, additional structure will be introduced into the global spectrum, making the polynomial fitting algorithm perform poorly. In this paper, we introduce an improved polynomial fitting algorithm - the Vari-Zeroth-Order Polynomial (VZOP) fitting and use it to fit the simulation data. This algorithm is developed for the upcoming Low-frequency Anechoic Chamber Experiment (LACE), yet it is a general method suitable for application in any single antenna-based global 21 cm signal experiment. VZOP defines a 24-hour averaged beam model that brings information about the antenna beam into the polynomial model. Assuming that the beam can be measured, VZOP can successfully recover the 21 cm absorption feature, even if the beam is extremely frequency-dependent. In real observations, due to various systematics, the corrected measured beam contains residual errors that are not completely random. Assuming the errors are frequency-dependent, VZOP is capable of recovering the 21 cm absorption feature even when the error reaches 10%. Even in the most extreme scenario where the errors are completely random, VZOP can at least give a fitting result that is not worse than the common polynomial fitting. In conclusion, the fitting effect of VZOP depends on the structure of the error and the accuracy of the beam measurement., Comment: 14 pages, 15 figures, Accepted for publication in MNRAS

Factuality is a crucial requirement in information seeking dialogue: the system should respond to the user's queries so that the responses are meaningful and aligned with the knowledge provided to the system. However, most modern large language models suffer from hallucinations, that is, they generate responses not supported by or contradicting the knowledge source. To mitigate the issue and increase faithfulness of information-seeking dialogue systems, we introduce BeInfo, a simple yet effective method that applies behavioural tuning to aid information-seeking dialogue. Relying on three standard datasets, we show that models tuned with BeInfo} become considerably more faithful to the knowledge source both for datasets and domains seen during BeInfo-tuning, as well as on unseen domains, when applied in a zero-shot manner. In addition, we show that the models with 3B parameters (e.g., Flan-T5) tuned with BeInfo demonstrate strong performance on data from real `production' conversations and outperform GPT4 when tuned on a limited amount of such realistic in-domain dialogues.

A crucial reason for the success of existing NeRF-based methods is to build a neural density field for the geometry representation via multiple perceptron layers (MLPs). MLPs are continuous functions, however, real geometry or density field is frequently discontinuous at the interface between the air and the surface. Such a contrary brings the problem of unfaithful geometry representation. To this end, this paper proposes spiking NeRF, which leverages spiking neurons and a hybrid Artificial Neural Network (ANN)-Spiking Neural Network (SNN) framework to build a discontinuous density field for faithful geometry representation. Specifically, we first demonstrate the reason why continuous density fields will bring inaccuracy. Then, we propose to use the spiking neurons to build a discontinuous density field. We conduct a comprehensive analysis for the problem of existing spiking neuron models and then provide the numerical relationship between the parameter of the spiking neuron and the theoretical accuracy of geometry. Based on this, we propose a bounded spiking neuron to build the discontinuous density field. Our method achieves SOTA performance. The source code and the supplementary material are available at https://github.com/liaozhanfeng/Spiking-NeRF.

We calculate the reflection of diffuse galactic emission by meteor trails and investigate its potential relationship to Meteor Radio Afterglow (MRA). The formula to calculate the reflection of diffuse galactic emission is derived from a simplified case, assuming that the signals are mirrored by the cylindrical over-dense ionization trail of meteors. The overall observed reflection is simulated through a ray tracing algorithm together with the diffuse galactic emission modelled by the GSM sky model. We demonstrate that the spectrum of the reflected signal is broadband and follows a power law with a negative spectral index of around -1.3. The intensity of the reflected signal varies with local sidereal time and the brightness of the meteor and can reach 2000 Jy. These results agree with some previous observations of MRAs. Therefore, we think that the reflection of galactic emission by meteor trails can be a possible mechanism causing MRAs, which is worthy of further research., Comment: 15 pages, 10 figures, 2 tables, accepted for publication in MNRAS, 10.1093/mnras/stad3429

Ad hoc teamwork requires an agent to cooperate with unknown teammates without prior coordination. Many works propose to abstract teammate instances into high-level representation of types and then pre-train the best response for each type. However, most of them do not consider the distribution of teammate instances within a type. This could expose the agent to the hidden risk of \emph{type confounding}. In the worst case, the best response for an abstract teammate type could be the worst response for all specific instances of that type. This work addresses the issue from the lens of causal inference. We first theoretically demonstrate that this phenomenon is due to the spurious correlation brought by uncontrolled teammate distribution. Then, we propose our solution, CTCAT, which disentangles such correlation through an instance-wise teammate feedback rectification. This operation reweights the interaction of teammate instances within a shared type to reduce the influence of type confounding. The effect of CTCAT is evaluated in multiple domains, including classic ad hoc teamwork tasks and real-world scenarios. Results show that CTCAT is robust to the influence of type confounding, a practical issue that directly hazards the robustness of our trained agents but was unnoticed in previous works., Comment: Accepted by ICML 2023

Spiking neural networks (SNNs) have shown advantages in computation and energy efficiency over traditional artificial neural networks (ANNs) thanks to their event-driven representations. SNNs also replace weight multiplications in ANNs with additions, which are more energy-efficient and less computationally intensive. However, it remains a challenge to train deep SNNs due to the discrete spike function. A popular approach to circumvent this challenge is ANN-to-SNN conversion. However, due to the quantization error and accumulating error, it often requires lots of time steps (high inference latency) to achieve high performance, which negates SNN's advantages. To this end, this paper proposes Fast-SNN that achieves high performance with low latency. We demonstrate the equivalent mapping between temporal quantization in SNNs and spatial quantization in ANNs, based on which the minimization of the quantization error is transferred to quantized ANN training. With the minimization of the quantization error, we show that the sequential error is the primary cause of the accumulating error, which is addressed by introducing a signed IF neuron model and a layer-wise fine-tuning mechanism. Our method achieves state-of-the-art performance and low latency on various computer vision tasks, including image classification, object detection, and semantic segmentation. Codes are available at: https://github.com/yangfan-hu/Fast-SNN., Comment: Accepted to IEEE Transactions on Pattern Analysis and Machine Intelligence

Uncalibrated photometric stereo (UPS) is challenging due to the inherent ambiguity brought by the unknown light. Although the ambiguity is alleviated on non-Lambertian objects, the problem is still difficult to solve for more general objects with complex shapes introducing irregular shadows and general materials with complex reflectance like anisotropic reflectance. To exploit cues from shadow and reflectance to solve UPS and improve performance on general materials, we propose DANI-Net, an inverse rendering framework with differentiable shadow handling and anisotropic reflectance modeling. Unlike most previous methods that use non-differentiable shadow maps and assume isotropic material, our network benefits from cues of shadow and anisotropic reflectance through two differentiable paths. Experiments on multiple real-world datasets demonstrate our superior and robust performance., Comment: Accepted by CVPR 2023

The ChatGPT, a lite and conversational variant of Generative Pretrained Transformer 4 (GPT-4) developed by OpenAI, is one of the milestone Large Language Models (LLMs) with billions of parameters. LLMs have stirred up much interest among researchers and practitioners in their impressive skills in natural language processing tasks, which profoundly impact various fields. This paper mainly discusses the future applications of LLMs in dentistry. We introduce two primary LLM deployment methods in dentistry, including automated dental diagnosis and cross-modal dental diagnosis, and examine their potential applications. Especially, equipped with a cross-modal encoder, a single LLM can manage multi-source data and conduct advanced natural language reasoning to perform complex clinical operations. We also present cases to demonstrate the potential of a fully automatic Multi-Modal LLM AI system for dentistry clinical application. While LLMs offer significant potential benefits, the challenges, such as data privacy, data quality, and model bias, need further study. Overall, LLMs have the potential to revolutionize dental diagnosis and treatment, which indicates a promising avenue for clinical application and research in dentistry.

Simulating the effects of skincare products on face is a potential new way to communicate the efficacy of skincare products in skin diagnostics and product recommendations. Furthermore, such simulations enable one to anticipate his/her skin conditions and better manage skin health. However, there is a lack of effective simulations today. In this paper, we propose the first simulation model to reveal facial pore changes after using skincare products. Our simulation pipeline consists of 2 steps: training data establishment and facial pore simulation. To establish training data, we collect face images with various pore quality indexes from short-term (8-weeks) clinical studies. People often experience significant skin fluctuations (due to natural rhythms, external stressors, etc.,), which introduces large perturbations in clinical data. To address this problem, we propose a sliding window mechanism to clean data and select representative index(es) to represent facial pore changes. Facial pore simulation stage consists of 3 modules: UNet-based segmentation module to localize facial pores; regression module to predict time-dependent warping hyperparameters; and deformation module, taking warping hyperparameters and pore segmentation labels as inputs, to precisely deform pores accordingly. The proposed simulation is able to render realistic facial pore changes. And this work will pave the way for future research in facial skin simulation and skincare product developments., Comment: 6 pages, 7 figures

The global neutral hydrogen 21 cm signal extracted from the all-sky averaged radio spectra is one of the signatures of the Cosmic Dawn and Epoch of Reionization (CD/EoR). The frequency-dependency of antenna beam patterns coupled with the strong foreground emission could introduce artificial spectral structures and cause false detection. A digital beamforming array could be potentially employed to form achromatic station beam patterns to overcome this problem. In this work, we discuss the method of forming achromatic beam patterns with a dense regular beamforming array to detect the global CD/EoR signal, covering topics including the array configuration, antenna weight optimization, and error estimation. We also show that based on the equivalence between a beamforming array and an interferometer, most antennas in the array can be removed by canceling redundant baselines. We present an example array design, optimize the antenna weights, and show the final array configuration by canceling redundant baselines. The performance of the example array is evaluated based on a simulation, which provides a positive indication towards the feasibility of detecting the CD/EoR signal using a dense digital beamforming array., Comment: 11 pages, 8 figures, submitted to MNRAS

Uncalibrated photometric stereo (UPS) is challenging due to the inherent ambiguity brought by unknown light. Existing solutions alleviate the ambiguity by either explicitly associating reflectance to light conditions or resolving light conditions in a supervised manner. This paper establishes an implicit relation between light clues and light estimation and solves UPS in an unsupervised manner. The key idea is to represent the reflectance as four neural intrinsics fields, i.e., position, light, specular, and shadow, based on which the neural light field is implicitly associated with light clues of specular reflectance and cast shadow. The unsupervised, joint optimization of neural intrinsics fields can be free from training data bias as well as accumulating error, and fully exploits all observed pixel values for UPS. Our method achieves a superior performance advantage over state-of-the-art UPS methods on public and self-collected datasets, under regular and challenging setups. The code will be released soon.

The shape of the faint-end of the high-z galaxy luminosity function (LF) informs early star formation and reionization physics during the Cosmic Dawn and Epoch of Reionization. Until recently, based on the strong gravitational lensing cluster deep surveys, the Hubble Frontier Fields (HFF) has found a potential turnover in the ultraviolet (UV) LF at z$\sim$6. In this paper, we analyze the contribution of extremely faint galaxies with the magnitude larger than the turnover magnitude in LF to cosmic reionization. We apply the measurement from HFF to our suppressed star formation efficiency model, including three free parameters: halo mass threshold $M_t$, curvature parameter $\beta$ and a UV conversion factor $l_{\rm UV}$. According to our fit of 68\% confidence level, the high-redshift star formation in haloes smaller than $ M_t=1.82^{+2.86}_{-1.08}\times10^{10} \rm M_{\odot}$ is found to be dampened. The turnover magnitude $\rm \gtrsim -13.99-2.45$, correspondingly the halo mass $\lesssim(4.57+20.03)\times10^{9} \rm M_{\odot}$. We find that the absorption trough in the global 21-cm signal is sensitive to our SFE model parameters. Together with ($\beta$, $l_{\rm UV}$) = ($2.17^{+2.42}_{-1.72}$, $9.33^{+0.43}_{-0.42} \rm ~erg~yr ~s^{-1}M_{\odot}^{-1})$, the trough locates at $\sim$ $134^{+10}_{-17}$ $\rm MHz$ with an amplitude of $\sim$ $-237^{-6}_{+7}$ $\rm mK$, compared to (106\rm MHz, -212\rm mK) in the absence of turnover. Besides, we find that the star formation of faint galaxies has also an impact on the 21-cm power spectra. The best fitting peak power decreases by $\sim4\%$ and shifts towards smaller scales from $0.88 h \rm Mpc^{-1}$ to $0.91 h \rm Mpc^{-1}$. According to our calculation, such impact is distinguishable with the forthcoming Square Kilometre Array., Comment: 11 pages, 5 figures; Accepted by MNRAS

Spectrum graph theory not only facilitate comprehensively reflect the topological structure and dynamic characteristics of networks, but also offer significant and noteworthy applications in theoretical chemistry, network science and other fields. Let $L_{n}^{8,4}$ represent a linear octagonal-quadrilateral network, consisting of $n$ eight-member ring and $n$ four-member ring. The M\"{o}bius graph $Q_{n}(8,4)$ is constructed by reverse identifying the opposite edges, whereas cylinder graph $Q'_{n}(8,4)$ identifies the opposite edges by order. In this paper, the explicit formulas of Kirchhoff indices and complexity of $Q_{n}(8,4)$ and $Q'_{n}(8,4)$ are demonstrated by Laplacian characteristic polynomials according to decomposition theorem and Vieta's theorem. In surprise, the Kirchhoff index of $Q_{n}(8,4)$($Q'_{n}(8,4)$) is approximately one-third half of its Wiener index as $n\to\infty$.

Recent advances in deep reinforcement learning (DRL) have largely promoted the performance of adaptive traffic signal control (ATSC). Nevertheless, regarding the implementation, most works are cumbersome in terms of storage and computation. This hinders their deployment on scenarios where resources are limited. In this work, we propose TinyLight, the first DRL-based ATSC model that is designed for devices with extremely limited resources. TinyLight first constructs a super-graph to associate a rich set of candidate features with a group of light-weighted network blocks. Then, to diminish the model's resource consumption, we ablate edges in the super-graph automatically with a novel entropy-minimized objective function. This enables TinyLight to work on a standalone microcontroller with merely 2KB RAM and 32KB ROM. We evaluate TinyLight on multiple road networks with real-world traffic demands. Experiments show that even with extremely limited resources, TinyLight still achieves competitive performance. The source code and appendix of this work can be found at \url{https://bit.ly/38hH8t8}., Comment: Accepted by IJCAI 2022 (Long Oral)