We provide estimates on the fat-shattering dimension of aggregation rules of real-valued function classes. The latter consists of all ways of choosing $k$ functions, one from each of the $k$ classes, and computing a pointwise function of them, such as the median, mean, and maximum. The bound is stated in terms of the fat-shattering dimensions of the component classes. For linear and affine function classes, we provide a considerably sharper upper bound and a matching lower bound, achieving, in particular, an optimal dependence on $k$. Along the way, we improve several known results in addition to pointing out and correcting a number of erroneous claims in the literature.
相關內容
Deep-learning models have been successful in biomedical image segmentation. To generalize for real-world deployment, test-time augmentation (TTA) methods are often used to transform the test image into different versions that are hopefully closer to the training domain. Unfortunately, due to the vast diversity of instance scale and image styles, many augmented test images produce undesirable results, thus lowering the overall performance. This work proposes a new TTA framework, S$^3$-TTA, which selects the suitable image scale and style for each test image based on a transformation consistency metric. In addition, S$^3$-TTA constructs an end-to-end augmentation-segmentation joint-training pipeline to ensure a task-oriented augmentation. On public benchmarks for cell and lung segmentation, S$^3$-TTA demonstrates improvements over the prior art by 3.4% and 1.3%, respectively, by simply augmenting the input data in testing phase.
Recent progresses on self-supervised 3D human action representation learning are largely attributed to contrastive learning. However, in conventional contrastive frameworks, the rich complementarity between different skeleton modalities remains under-explored. Moreover, optimized with distinguishing self-augmented samples, models struggle with numerous similar positive instances in the case of limited action categories. In this work, we tackle the aforementioned problems by introducing a general Inter- and Intra-modal Mutual Distillation (I$^2$MD) framework. In I$^2$MD, we first re-formulate the cross-modal interaction as a Cross-modal Mutual Distillation (CMD) process. Different from existing distillation solutions that transfer the knowledge of a pre-trained and fixed teacher to the student, in CMD, the knowledge is continuously updated and bidirectionally distilled between modalities during pre-training. To alleviate the interference of similar samples and exploit their underlying contexts, we further design the Intra-modal Mutual Distillation (IMD) strategy, In IMD, the Dynamic Neighbors Aggregation (DNA) mechanism is first introduced, where an additional cluster-level discrimination branch is instantiated in each modality. It adaptively aggregates highly-correlated neighboring features, forming local cluster-level contrasting. Mutual distillation is then performed between the two branches for cross-level knowledge exchange. Extensive experiments on three datasets show that our approach sets a series of new records.
For a hypergraph $H$, the transversal is a subset of vertices whose intersection with every edge is nonempty. The cardinality of a minimum transversal is the transversal number of $H$, denoted by $\tau(H)$. The Tuza constant $c_k$ is defined as $\sup{\tau(H)/ (m+n)}$, where $H$ ranges over all $k$-uniform hypergraphs, with $m$ and $n$ being the number of edges and vertices, respectively. We give an upper bound and a lower bound on $c_k$. The upper bound improves the known ones for $k\geq 7$, and the lower bound improves the known ones for $k\in\{7, 8, 10, 11, 13, 14, 17\}$.
The monotonicity of discrete Laplacian implies discrete maximum principle, which in general does not hold for high order schemes. The $Q^2$ spectral element method has been proven monotone on a uniform rectangular mesh. In this paper we prove the monotonicity of the $Q^2$ spectral element method on quasi-uniform rectangular meshes under certain mesh constraints. In particular, we propose a relaxed Lorenz's condition for proving monotonicity.
Increases in the deployment of machine learning algorithms for applications that deal with sensitive data have brought attention to the issue of fairness in machine learning. Many works have been devoted to applications that require different demographic groups to be treated fairly. However, algorithms that aim to satisfy inter-group fairness (also called group fairness) may inadvertently treat individuals within the same demographic group unfairly. To address this issue, we introduce a formal definition of within-group fairness that maintains fairness among individuals from within the same group. We propose a pre-processing framework to meet both inter- and within-group fairness criteria with little compromise in accuracy. The framework maps the feature vectors of members from different groups to an inter-group-fair canonical domain before feeding them into a scoring function. The mapping is constructed to preserve the relative relationship between the scores obtained from the unprocessed feature vectors of individuals from the same demographic group, guaranteeing within-group fairness. We apply this framework to the COMPAS risk assessment and Law School datasets and compare its performance in achieving inter-group and within-group fairness to two regularization-based methods.
Current approaches in paraphrase generation and detection heavily rely on a single general similarity score, ignoring the intricate linguistic properties of language. This paper introduces two new tasks to address this shortcoming by considering paraphrase types - specific linguistic perturbations at particular text positions. We name these tasks Paraphrase Type Generation and Paraphrase Type Detection. Our results suggest that while current techniques perform well in a binary classification scenario, i.e., paraphrased or not, the inclusion of fine-grained paraphrase types poses a significant challenge. While most approaches are good at generating and detecting general semantic similar content, they fail to understand the intrinsic linguistic variables they manipulate. Models trained in generating and identifying paraphrase types also show improvements in tasks without them. In addition, scaling these models further improves their ability to understand paraphrase types. We believe paraphrase types can unlock a new paradigm for developing paraphrase models and solving tasks in the future.
Simultaneous machine translation (SiMT) requires a robust read/write policy in conjunction with a high-quality translation model. Traditional methods rely on either a fixed wait-$k$ policy coupled with a standalone wait-$k$ translation model, or an adaptive policy jointly trained with the translation model. In this study, we propose a more flexible approach by decoupling the adaptive policy model from the translation model. Our motivation stems from the observation that a standalone multi-path wait-$k$ model performs competitively with adaptive policies utilized in state-of-the-art SiMT approaches. Specifically, we introduce DaP, a divergence-based adaptive policy, that makes read/write decisions for any translation model based on the potential divergence in translation distributions resulting from future information. DaP extends a frozen wait-$k$ model with lightweight parameters, and is both memory and computation efficient. Experimental results across various benchmarks demonstrate that our approach offers an improved trade-off between translation accuracy and latency, outperforming strong baselines.
We settle the parameterized complexities of several variants of independent set reconfiguration and dominating set reconfiguration, parameterized by the number of tokens. We show that both problems are XL-complete when there is no limit on the number of moves, XNL-complete when a maximum length $\ell$ for the sequence is given in binary in the input, and XNLP-complete when $\ell$ is given in unary. The problems were known to be $\mathrm{W}[1]$- and $\mathrm{W}[2]$-hard respectively when $\ell$ is also a parameter. We complete the picture by showing membership in those classes. Moreover, we show that for all the variants that we consider, token sliding and token jumping are equivalent under pl-reductions. We introduce partitioned variants of token jumping and token sliding, and give pl-reductions between the four variants that have precise control over the number of tokens and the length of the reconfiguration sequence.
Byzantine agreement (BA), the task of $n$ parties to agree on one of their input bits in the face of malicious agents, is a powerful primitive that lies at the core of a vast range of distributed protocols. Interestingly, in protocols with the best overall communication, the demands of the parties are highly unbalanced: the amortized cost is $\tilde O(1)$ bits per party, but some parties must send $\Omega(n)$ bits. In best known balanced protocols, the overall communication is sub-optimal, with each party communicating $\tilde O(\sqrt{n})$. In this work, we ask whether asymmetry is inherent for optimizing total communication. Our contributions in this line are as follows: 1) We define a cryptographic primitive, succinctly reconstructed distributed signatures (SRDS), that suffices for constructing $\tilde O(1)$ balanced BA. We provide two constructions of SRDS from different cryptographic and Public-Key Infrastructure (PKI) assumptions. 2) The SRDS-based BA follows a paradigm of boosting from "almost-everywhere" agreement to full agreement, and does so in a single round. We prove that PKI setup and cryptographic assumptions are necessary for such protocols in which every party sends $o(n)$ messages. 3) We further explore connections between a natural approach toward attaining SRDS and average-case succinct non-interactive argument systems (SNARGs) for a particular type of NP-Complete problems (generalizing Subset-Sum and Subset-Product). Our results provide new approaches forward, as well as limitations and barriers, towards minimizing per-party communication of BA. In particular, we construct the first two BA protocols with $\tilde O(1)$ balanced communication, offering a tradeoff between setup and cryptographic assumptions, and answering an open question presented by King and Saia (DISC'09).
Most existing works in visual question answering (VQA) are dedicated to improving the accuracy of predicted answers, while disregarding the explanations. We argue that the explanation for an answer is of the same or even more importance compared with the answer itself, since it makes the question and answering process more understandable and traceable. To this end, we propose a new task of VQA-E (VQA with Explanation), where the computational models are required to generate an explanation with the predicted answer. We first construct a new dataset, and then frame the VQA-E problem in a multi-task learning architecture. Our VQA-E dataset is automatically derived from the VQA v2 dataset by intelligently exploiting the available captions. We have conducted a user study to validate the quality of explanations synthesized by our method. We quantitatively show that the additional supervision from explanations can not only produce insightful textual sentences to justify the answers, but also improve the performance of answer prediction. Our model outperforms the state-of-the-art methods by a clear margin on the VQA v2 dataset.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191