R^2VOS: Robust Referring Video Object Segmentation via Relational Multimodal Cycle Consistency

Referring video object segmentation (R-VOS) aims to segment the object masks in a video given a referring linguistic expression to the object. It is a recently introduced task attracting growing research attention. However, all existing works make a strong assumption: The object depicted by the expression must exist in the video, namely, the expression and video must have an object-level semantic consensus. This is often violated in real-world applications where an expression can be queried to false videos, and existing methods always fail in such false queries due to abusing the assumption. In this work, we emphasize that studying semantic consensus is necessary to improve the robustness of R-VOS. Accordingly, we pose an extended task from R-VOS without the semantic consensus assumption, named Robust R-VOS ($\mathrm{R}^2$-VOS). The $\mathrm{R}^2$-VOS task is essentially related to the joint modeling of the primary R-VOS task and its dual problem (text reconstruction). We embrace the observation that the embedding spaces have relational consistency through the cycle of text-video-text transformation, which connects the primary and dual problems. We leverage the cycle consistency to discriminate the semantic consensus, thus advancing the primary task. Parallel optimization of the primary and dual problems are enabled by introducing an early grounding medium. A new evaluation dataset, $\mathrm{R}^2$-Youtube-VOS, is collected to measure the robustness of R-VOS models against unpaired videos and expressions. Extensive experiments demonstrate that our method not only identifies negative pairs of unrelated expressions and videos, but also improves the segmentation accuracy for positive pairs with a superior disambiguating ability. Our model achieves the state-of-the-art performance on Ref-DAVIS17, Ref-Youtube-VOS, and the novel $\mathrm{R}^2$-Youtube-VOS dataset.

AI for CSI Feedback Enhancement in 5G-Advanced and 6G

The 3rd Generation Partnership Project has started the study of Release 18 in 2021. Artificial intelligence (AI)-native air interface is one of the key features of Release 18, where AI for channel state information (CSI) feedback enhancement is selected as the representative use case. This article provides a comprehensive overview of AI for CSI feedback enhancement in 5G-Advanced and 6G. The scope of the AI for CSI feedback enhancement in 5G-Advanced, including overhead reduction, accuracy improvement, and channel prediction, is first presented and discussed. Then, three representative frameworks of AI-enabled CSI feedback, including one-sided implicit feedback, two-sided autoencoder-based implicit feedback, and two-sided explicit feedback, are introduced and compared. Finally, the considerations in the standardization of AI for CSI feedback enhancement, especially focusing on evaluation, complexity, collaboration, generalization, information sharing, joint design with channel prediction, and reciprocity, have been identified and discussed. This article provides a guideline for the standardization study of the AI-based CSI feedback enhancement.

Randomized Coordinate Subgradient Method for Nonsmooth Optimization

Nonsmooth optimization finds wide applications in many engineering fields. In this work, we propose to utilize the {Randomized Coordinate Subgradient Method} (RCS) for solving both nonsmooth convex and nonsmooth nonconvex (nonsmooth weakly convex) optimization problems. At each iteration, RCS randomly selects one block coordinate rather than all the coordinates to update. Motivated by practical applications, we consider the {linearly bounded subgradients assumption} for the objective function, which is much more general than the Lipschitz continuity assumption. Under such a general assumption, we conduct thorough convergence analysis for RCS in both convex and nonconvex cases and establish both expected convergence rate and almost sure asymptotic convergence results. In order to derive these convergence results, we establish a convergence lemma and the relationship between the global metric subregularity properties of a weakly convex function and its Moreau envelope, which are fundamental and of independent interests. Finally, we conduct several experiments to show the possible superiority of RCS over the subgradient method.

Finite-Time Analysis of Fully Decentralized Single-Timescale Actor-Critic

Decentralized Actor-Critic (AC) algorithms have been widely utilized for multi-agent reinforcement learning (MARL) and have achieved remarkable success. Apart from its empirical success, the theoretical convergence property of decentralized AC algorithms is largely unexplored. The existing finite-time convergence results are derived based on either double-loop update or two-timescale step sizes rule, which is not often adopted in real implementation. In this work, we introduce a fully decentralized AC algorithm, where actor, critic, and global reward estimator are updated in an alternating manner with step sizes being of the same order, namely, we adopt the \emph{single-timescale} update. Theoretically, using linear approximation for value and reward estimation, we show that our algorithm has sample complexity of $\tilde{\mathcal{O}}(\epsilon^{-2})$ under Markovian sampling, which matches the optimal complexity with double-loop implementation (here, $\tilde{\mathcal{O}}$ hides a log term). The sample complexity can be improved to ${\mathcal{O}}(\epsilon^{-2})$ under the i.i.d. sampling scheme. The central to establishing our complexity results is \emph{the hidden smoothness of the optimal critic variable} we revealed. We also provide a local action privacy-preserving version of our algorithm and its analysis. Finally, we conduct experiments to show the superiority of our algorithm over the existing decentralized AC algorithms.

A Unified Convergence Theorem for Stochastic Optimization Methods

In this work, we provide a fundamental unified convergence theorem used for deriving expected and almost sure convergence results for a series of stochastic optimization methods. Our unified theorem only requires to verify several representative conditions and is not tailored to any specific algorithm. As a direct application, we recover expected and almost sure convergence results of the stochastic gradient method (SGD) and random reshuffling (RR) under more general settings. Moreover, we establish new expected and almost sure convergence results for the stochastic proximal gradient method (prox-SGD) and stochastic model-based methods (SMM) for nonsmooth nonconvex optimization problems. These applications reveal that our unified theorem provides a plugin-type convergence analysis and strong convergence guarantees for a wide class of stochastic optimization methods.

AdaLoGN: Adaptive Logic Graph Network for Reasoning-Based Machine Reading Comprehension

Recent machine reading comprehension datasets such as ReClor and LogiQA require performing logical reasoning over text. Conventional neural models are insufficient for logical reasoning, while symbolic reasoners cannot directly apply to text. To meet the challenge, we present a neural-symbolic approach which, to predict an answer, passes messages over a graph representing logical relations between text units. It incorporates an adaptive logic graph network (AdaLoGN) which adaptively infers logical relations to extend the graph and, essentially, realizes mutual and iterative reinforcement between neural and symbolic reasoning. We also implement a novel subgraph-to-node message passing mechanism to enhance context-option interaction for answering multiple-choice questions. Our approach shows promising results on ReClor and LogiQA.

On the Optimization Landscape of Neural Collapse under MSE Loss: Global Optimality with Unconstrained Features

When training deep neural networks for classification tasks, an intriguing empirical phenomenon has been widely observed in the last-layer classifiers and features, where (i) the class means and the last-layer classifiers all collapse to the vertices of a Simplex Equiangular Tight Frame (ETF) up to scaling, and (ii) cross-example within-class variability of last-layer activations collapses to zero. This phenomenon is called Neural Collapse (NC), which seems to take place regardless of the choice of loss functions. In this work, we justify NC under the mean squared error (MSE) loss, where recent empirical evidence shows that it performs comparably or even better than the de-facto cross-entropy loss. Under a simplified unconstrained feature model, we provide the first global landscape analysis for vanilla nonconvex MSE loss and show that the (only!) global minimizers are neural collapse solutions, while all other critical points are strict saddles whose Hessian exhibit negative curvature directions. Furthermore, we justify the usage of rescaled MSE loss by probing the optimization landscape around the NC solutions, showing that the landscape can be improved by tuning the rescaling hyperparameters. Finally, our theoretical findings are experimentally verified on practical network architectures.

Distributed Random Reshuffling over Networks

In this paper, we consider the distributed optimization problem where $n$ agents, each possessing a local cost function, collaboratively minimize the average of the local cost functions over a connected network. To solve the problem, we propose a distributed random reshuffling (D-RR) algorithm that combines the classical distributed gradient descent (DGD) method and Random Reshuffling (RR). We show that D-RR inherits the superiority of RR for both smooth strongly convex and smooth nonconvex objective functions. In particular, for smooth strongly convex objective functions, D-RR achieves $\mathcal{O}(1/T^2)$ rate of convergence (here, $T$ counts the total number of iterations) in terms of the squared distance between the iterate and the unique minimizer. When the objective function is assumed to be smooth nonconvex and has Lipschitz continuous component functions, we show that D-RR drives the squared norm of gradient to $0$ at a rate of $\mathcal{O}(1/T^{2/3})$. These convergence results match those of centralized RR (up to constant factors).