Neural scene representations, both continuous and discrete, have recently emerged as a powerful new paradigm for 3D scene understanding. Recent efforts have tackled unsupervised discovery of object-centric neural scene representations. However, the high cost of ray-marching, exacerbated by the fact that each object representation has to be ray-marched separately, leads to insufficiently sampled radiance fields and thus, noisy renderings, poor framerates, and high memory and time complexity during training and rendering. Here, we propose to represent objects in an object-centric, compositional scene representation as light fields. We propose a novel light field compositor module that enables reconstructing the global light field from a set of object-centric light fields. Dubbed Compositional Object Light Fields (COLF), our method enables unsupervised learning of object-centric neural scene representations, state-of-the-art reconstruction and novel view synthesis performance on standard datasets, and rendering and training speeds at orders of magnitude faster than existing 3D approaches.
相關內容
Not all camouflages are equally effective, as even a partially visible contour or a slight color difference can make the animal stand out and break its camouflage. In this paper, we address the question of what makes a camouflage successful, by proposing three scores for automatically assessing its effectiveness. In particular, we show that camouflage can be measured by the similarity between background and foreground features and boundary visibility. We use these camouflage scores to assess and compare all available camouflage datasets. We also incorporate the proposed camouflage score into a generative model as an auxiliary loss and show that effective camouflage images or videos can be synthesised in a scalable manner. The generated synthetic dataset is used to train a transformer-based model for segmenting camouflaged animals in videos. Experimentally, we demonstrate state-of-the-art camouflage breaking performance on the public MoCA-Mask benchmark.
We study stochastic Cubic Newton methods for solving general possibly non-convex minimization problems. We propose a new framework, which we call the helper framework, that provides a unified view of the stochastic and variance-reduced second-order algorithms equipped with global complexity guarantees. It can also be applied to learning with auxiliary information. Our helper framework offers the algorithm designer high flexibility for constructing and analyzing the stochastic Cubic Newton methods, allowing arbitrary size batches, and the use of noisy and possibly biased estimates of the gradients and Hessians, incorporating both the variance reduction and the lazy Hessian updates. We recover the best-known complexities for the stochastic and variance-reduced Cubic Newton, under weak assumptions on the noise. A direct consequence of our theory is the new lazy stochastic second-order method, which significantly improves the arithmetic complexity for large dimension problems. We also establish complexity bounds for the classes of gradient-dominated objectives, that include convex and strongly convex problems. For Auxiliary Learning, we show that using a helper (auxiliary function) can outperform training alone if a given similarity measure is small.
We provide a method, based on automata theory, to mechanically prove the correctness of many numeration systems based on Fibonacci numbers. With it, long case-based and induction-based proofs of correctness can be replaced by simply constructing a regular expression (or finite automaton) specifying the rules for valid representations, followed by a short computation. Examples of the systems that can be handled using our technique include Brown's lazy representation (1965), the far-difference representation developed by Alpert (2009), and three representations proposed by Hajnal (2023). We also provide three additional systems and prove their validity.
Subsethood, which is to measure the degree of set inclusion relation, is predominant in fuzzy set theory. This paper introduces some basic concepts of spatial granules, coarse-fine relation, and operations like meet, join, quotient meet and quotient join. All the atomic granules can be hierarchized by set-inclusion relation and all the granules can be hierarchized by coarse-fine relation. Viewing an information system from the micro and the macro perspectives, we can get a micro knowledge space and a micro knowledge space, from which a rough set model and a spatial rough granule model are respectively obtained. The classical rough set model is the special case of the rough set model induced from the micro knowledge space, while the spatial rough granule model will be play a pivotal role in the problem-solving of structures. We discuss twelve axioms of monotone increasing subsethood and twelve corresponding axioms of monotone decreasing supsethood, and generalize subsethood and supsethood to conditional granularity and conditional fineness respectively. We develop five conditional granularity measures and five conditional fineness measures and prove that each conditional granularity or fineness measure satisfies its corresponding twelve axioms although its subsethood or supsethood measure only hold one of the two boundary conditions. We further define five conditional granularity entropies and five conditional fineness entropies respectively, and each entropy only satisfies part of the boundary conditions but all the ten monotone conditions.
This paper begins with a description of methods for estimating probability density functions for images that reflects the observation that such data is usually constrained to lie in restricted regions of the high-dimensional image space - not every pattern of pixels is an image. It is common to say that images lie on a lower-dimensional manifold in the high-dimensional space. However, although images may lie on such lower-dimensional manifolds, it is not the case that all points on the manifold have an equal probability of being images. Images are unevenly distributed on the manifold, and our task is to devise ways to model this distribution as a probability distribution. In pursuing this goal, we consider generative models that are popular in AI and computer vision community. For our purposes, generative/probabilistic models should have the properties of 1) sample generation: it should be possible to sample from this distribution according to the modelled density function, and 2) probability computation: given a previously unseen sample from the dataset of interest, one should be able to compute the probability of the sample, at least up to a normalising constant. To this end, we investigate the use of methods such as normalising flow and diffusion models. We then show that such probabilistic descriptions can be used to construct defences against adversarial attacks. In addition to describing the manifold in terms of density, we also consider how semantic interpretations can be used to describe points on the manifold. To this end, we consider an emergent language framework which makes use of variational encoders to produce a disentangled representation of points that reside on a given manifold. Trajectories between points on a manifold can then be described in terms of evolving semantic descriptions.
We consider outlier-robust and sparse estimation of linear regression coefficients, when the covariates and the noises are contaminated by adversarial outliers and noises are sampled from a heavy-tailed distribution. Our results present sharper error bounds under weaker assumptions than prior studies that share similar interests with this study. Our analysis relies on some sharp concentration inequalities resulting from generic chaining.
The rapid advancement of large language models, such as the Generative Pre-trained Transformer (GPT) series, has had significant implications across various disciplines. In this study, we investigate the potential of the state-of-the-art large language model (GPT-4) for planning tasks. We explore its effectiveness in multiple planning subfields, highlighting both its strengths and limitations. Through a comprehensive examination, we identify areas where large language models excel in solving planning problems and reveal the constraints that limit their applicability. Our empirical analysis focuses on GPT-4's performance in planning domain extraction, graph search path planning, and adversarial planning. We then propose a way of fine-tuning a domain-specific large language model to improve its Chain of Thought (CoT) capabilities for the above-mentioned tasks. The results provide valuable insights into the potential applications of large language models in the planning domain and pave the way for future research to overcome their limitations and expand their capabilities.
There have been recent advances in the analysis and visualization of 3D symmetric tensor fields, with a focus on the robust extraction of tensor field topology. However, topological features such as degenerate curves and neutral surfaces do not live in isolation. Instead, they intriguingly interact with each other. In this paper, we introduce the notion of {\em topological graph} for 3D symmetric tensor fields to facilitate global topological analysis of such fields. The nodes of the graph include degenerate curves and regions bounded by neutral surfaces in the domain. The edges in the graph denote the adjacency information between the regions and degenerate curves. In addition, we observe that a degenerate curve can be a loop and even a knot and that two degenerate curves (whether in the same region or not) can form a link. We provide a definition and theoretical analysis of individual degenerate curves in order to help understand why knots and links may occur. Moreover, we differentiate between wedges and trisectors, thus making the analysis more detailed about degenerate curves. We incorporate this information into the topological graph. Such a graph can not only reveal the global structure in a 3D symmetric tensor field but also allow two symmetric tensor fields to be compared. We demonstrate our approach by applying it to solid mechanics and material science data sets.
Large Language Models (LLMs) have shown excellent generalization capabilities that have led to the development of numerous models. These models propose various new architectures, tweaking existing architectures with refined training strategies, increasing context length, using high-quality training data, and increasing training time to outperform baselines. Analyzing new developments is crucial for identifying changes that enhance training stability and improve generalization in LLMs. This survey paper comprehensively analyses the LLMs architectures and their categorization, training strategies, training datasets, and performance evaluations and discusses future research directions. Moreover, the paper also discusses the basic building blocks and concepts behind LLMs, followed by a complete overview of LLMs, including their important features and functions. Finally, the paper summarizes significant findings from LLM research and consolidates essential architectural and training strategies for developing advanced LLMs. Given the continuous advancements in LLMs, we intend to regularly update this paper by incorporating new sections and featuring the latest LLM models.
The concept of causality plays an important role in human cognition . In the past few decades, causal inference has been well developed in many fields, such as computer science, medicine, economics, and education. With the advancement of deep learning techniques, it has been increasingly used in causal inference against counterfactual data. Typically, deep causal models map the characteristics of covariates to a representation space and then design various objective optimization functions to estimate counterfactual data unbiasedly based on the different optimization methods. This paper focuses on the survey of the deep causal models, and its core contributions are as follows: 1) we provide relevant metrics under multiple treatments and continuous-dose treatment; 2) we incorporate a comprehensive overview of deep causal models from both temporal development and method classification perspectives; 3) we assist a detailed and comprehensive classification and analysis of relevant datasets and source code.
小貼士
登錄享
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191