Automatically composing pop music with a satisfactory structure is an attractive but challenging topic. Although the musical structure is easy to be perceived by human, it is difficult to be described clearly and defined accurately. And it is still f
ar from being solved that how we should model the structure in pop music generation. In this paper, we propose to leverage harmony-aware learning for structure-enhanced pop music generation. On the one hand, one of the participants of harmony, chord, represents the harmonic set of multiple notes, which is integrated closely with the spatial structure of music, texture. On the other hand, the other participant of harmony, chord progression, usually accompanies with the development of the music, which promotes the temporal structure of music, form. Besides, when chords evolve into chord progression, the texture and the form can be bridged by the harmony naturally, which contributes to the joint learning of the two structures. Furthermore, we propose the Harmony-Aware Hierarchical Music Transformer (HAT), which can exploit the structure adaptively from the music, and interact on the music tokens at multiple levels to enhance the signals of the structure in various musical elements. Results of subjective and objective evaluations demonstrate that HAT significantly improves the quality of generated music, especially in the structureness.
In this paper, SAR image reconstruction with joint phase error estimation (autofocusing) is formulated as an inverse problem. An optimization model utilising a sparsity-enforcing Cauchy regularizer is proposed, and an alternating minimization framewo
rk is used to solve it, in which the desired image and the phase errors are optimized alternatively. For the image reconstruction sub-problem (f-sub-problem), two methods are presented capable of handling the problems complex nature, and we thus present two variants of our SAR image autofocusing algorithm. Firstly, we design a complex version of the forward-backward splitting algorithm (CFBA) to solve the f-sub-problem iteratively. For the second variant, the Wirtinger alternating minimization autofocusing (WAMA) method is presented, in which techniques of Wirtinger calculus are utilized to minimize the complex-valued cost function in the f-sub-problem in a direct fashion. For both methods, the phase error estimation sub-problem is solved by simply expanding and observing its cost function. Moreover, the convergence of both algorithms is discussed in detail. By conducting experiments on both simulated scenes and real SAR images, the proposed method is demonstrated to give impressive autofocusing results compared to other state of the art methods.
In the question answering(QA) task, multi-hop reasoning framework has been extensively studied in recent years to perform more efficient and interpretable answer reasoning on the Knowledge Graph(KG). However, multi-hop reasoning is inapplicable for a
nswering n-ary fact questions due to its linear reasoning nature. We discover that there are two feasible improvements: 1) upgrade the basic reasoning unit from entity or relation to fact; and 2) upgrade the reasoning structure from chain to tree. Based on these, we propose a novel fact-tree reasoning framework, through transforming the question into a fact tree and performing iterative fact reasoning on it to predict the correct answer. Through a comprehensive evaluation on the n-ary fact KGQA dataset introduced by this work, we demonstrate that the proposed fact-tree reasoning framework has the desired advantage of high answer prediction accuracy. In addition, we also evaluate the fact-tree reasoning framework on two binary KGQA datasets and show that our approach also has a strong reasoning ability compared with several excellent baselines. This work has direct implications for exploring complex reasoning scenarios and provides a preliminary baseline approach.
Graph convolutional networks (GCNs) have recently enabled a popular class of algorithms for collaborative filtering (CF). Nevertheless, the theoretical underpinnings of their empirical successes remain elusive. In this paper, we endeavor to obtain a
better understanding of GCN-based CF methods via the lens of graph signal processing. By identifying the critical role of smoothness, a key concept in graph signal processing, we develop a unified graph convolution-based framework for CF. We prove that many existing CF methods are special cases of this framework, including the neighborhood-based methods, low-rank matrix factorization, linear auto-encoders, and LightGCN, corresponding to different low-pass filters. Based on our framework, we then present a simple and computationally efficient CF baseline, which we shall refer to as Graph Filter based Collaborative Filtering (GF-CF). Given an implicit feedback matrix, GF-CF can be obtained in a closed form instead of expensive training with back-propagation. Experiments will show that GF-CF achieves competitive or better performance against deep learning-based methods on three well-known datasets, notably with a $70%$ performance gain over LightGCN on the Amazon-book dataset.
Transformers have improved the state-of-the-art across numerous tasks in sequence modeling. Besides the quadratic computational and memory complexity w.r.t the sequence length, the self-attention mechanism only processes information at the same scale
, i.e., all attention heads are in the same resolution, resulting in the limited power of the Transformer. To remedy this, we propose a novel and efficient structure named Adaptive Multi-Resolution Attention (AdaMRA for short), which scales linearly to sequence length in terms of time and space. Specifically, we leverage a multi-resolution multi-head attention mechanism, enabling attention heads to capture long-range contextual information in a coarse-to-fine fashion. Moreover, to capture the potential relations between query representation and clues of different attention granularities, we leave the decision of which resolution of attention to use to query, which further improves the models capacity compared to vanilla Transformer. In an effort to reduce complexity, we adopt kernel attention without degrading the performance. Extensive experiments on several benchmarks demonstrate the effectiveness and efficiency of our model by achieving a state-of-the-art performance-efficiency-memory trade-off. To facilitate AdaMRA utilization by the scientific community, the code implementation will be made publicly available.
Conditional average treatment effects (CATEs) allow us to understand the effect heterogeneity across a large population of individuals. However, typical CATE learners assume all confounding variables are measured in order for the CATE to be identifia
ble. Often, this requirement is satisfied by simply collecting many variables, at the expense of increased sample complexity for estimating CATEs. To combat this, we propose an energy-based model (EBM) that learns a low-dimensional representation of the variables by employing a noise contrastive loss function. With our EBM we introduce a preprocessing step that alleviates the dimensionality curse for any existing model and learner developed for estimating CATE. We prove that our EBM keeps the representations partially identifiable up to some universal constant, as well as having universal approximation capability to avoid excessive information loss from model misspecification; these properties combined with our loss function, enable the representations to converge and keep the CATE estimation consistent. Experiments demonstrate the convergence of the representations, as well as show that estimating CATEs on our representations performs better than on the variables or the representations obtained via various benchmark dimensionality reduction methods.
Selecting the most influential agent in a network has huge practical value in applications. However, in many scenarios, the graph structure can only be known from agents reports on their connections. In a self-interested setting, agents may strategic
ally hide some connections to make themselves seem to be more important. In this paper, we study the incentive compatible (IC) selection mechanism to prevent such manipulations. Specifically, we model the progeny of an agent as her influence power, i.e., the number of nodes in the subgraph rooted at her. We then propose the Geometric Mechanism, which selects an agent with at least 1/2 of the optimal progeny in expectation under the properties of incentive compatibility and fairness. Fairness requires that two roots with the same contribution in two graphs are assigned the same probability. Furthermore, we prove an upper bound of 1/(1+ln 2) for any incentive compatible and fair selection mechanisms.
Deep learning models are notoriously data-hungry. Thus, there is an urging need for data-efficient techniques in medical image analysis, where well-annotated data are costly and time consuming to collect. Motivated by the recently revived Copy-Paste
augmentation, we propose TumorCP, a simple but effective object-level data augmentation method tailored for tumor segmentation. TumorCP is online and stochastic, providing unlimited augmentation possibilities for tumors subjects, locations, appearances, as well as morphologies. Experiments on kidney tumor segmentation task demonstrate that TumorCP surpasses the strong baseline by a remarkable margin of 7.12% on tumor Dice. Moreover, together with image-level data augmentation, it beats the current state-of-the-art by 2.32% on tumor Dice. Comprehensive ablation studies are performed to validate the effectiveness of TumorCP. Meanwhile, we show that TumorCP can lead to striking improvements in extremely low-data regimes. Evaluated with only 10% labeled data, TumorCP significantly boosts tumor Dice by 21.87%. To the best of our knowledge, this is the very first work exploring and extending the Copy-Paste design in medical imaging domain. Code is available at: https://github.com/YaoZhang93/TumorCP.
Centralized monetary policy, leading to persistent inflation, is often inconsistent, untrustworthy, and unpredictable. Algorithmic stable coins enabled by blockchain technology are promising in solving this problem. Algorithmic stable coins utilize a
monetary policy that is entirely rule-based. However, there is little understanding about how to optimize the rule. We propose a model that trade-offs between the price and supply stability. We further study the comparative statistics by varying several design features. Finally, we discuss the empirical implications and further research for industry applications.
This paper presents an efficient approach for the conceptual design of architectural surfaces which are composed of triangular panels. Given an initial design, the proposed method finds a triangulated surface with user-specified Gaussian curvatures (
not limited to constant Gaussian curvatures) while keeping some of the vertices fixed. In addition, the conformal class of the final design can be specified; that is, the user has control over the shape (the corner angles) of each triangular panel. The panels could be encouraged to form a regular tessellation or kept close to those of the initial design. This allows the free-form design of discrete architectural surfaces that achive curvature requirements posed by stiffness and constructability. Furthermore, controllability on the conformal class suppresses possible distortion of the panels, resulting in higher structural performance and aesthetics. Our method relies on the idea in discrete differential geometry called circle packing. In this line of research, the discrete Ricci flow has been widely used for surface modelling. However, it is not trivial to incorporate constraints such as boundary locations and convexity of the spanned surface, which are essential to architectural applications. Due to this difficulty, few concrete applications of the discrete Ricci flow have been reported which specifically aims at creation of architectural surfaces. We propose a perturbation of the discrete Ricci energy and develop a least-squares-based optimisation scheme to address these problems with a working implementation available online.
