Recent state-of-the-art methods for neural architecture search (NAS) exploit gradient-based optimization by relaxing the problem into continuous optimization over architectures and shared-weights, a noisy process that remains poorly understood. We argue for the study of single-level empirical risk minimization to understand NAS with weight-sharing, reducing the design of NAS methods to devising optimizers and regularizers that can quickly obtain high-quality solutions to this problem. Invoking the theory of mirror descent, we present a geometry-aware framework that exploits the underlying structure of this optimization to return sparse architectural parameters, leading to simple yet novel algorithms that enjoy fast convergence guarantees and achieve state-of-the-art accuracy on the latest NAS benchmarks in computer vision. Notably, we exceed the best published results for both CIFAR and ImageNet on both the DARTS search space and NAS-Bench201; on the latter we achieve near-oracle-optimal performance on CIFAR-10 and CIFAR-100. Together, our theory and experiments demonstrate a principled way to co-design optimizers and continuous relaxations of discrete NAS search spaces.
Technion - Israel Institute of Technology（以色列理工学院）
Cloud computing is becoming increasingly popular as a platform for distributed training of deep neural networks. Synchronous stochastic gradient descent (SSGD) suffers from substantial slowdowns due to stragglers if the environment is non-dedicated, as is common in cloud computing. Asynchronous SGD (ASGD) methods are immune to these slowdowns but are scarcely used due to gradient staleness, which encumbers the convergence process. Recent techniques have had limited success mitigating the gradient staleness when scaling up to many workers (computing nodes). In this paper we define the Gap as a …
University of California, Irvine（加利福尼亚大学尔湾分校）；Department of Mathematics（数学系）
We propose an adaptive projection-gradient descent-shrinkage-splitting method (APGDSSM) to integrate penalty based channel pruning into quantization-aware training (QAT). APGDSSM concurrently searches weights in both the quantized subspace and the sparse subspace. APGDSSM uses shrinkage operator and a splitting technique to create sparse weights, as well as the Group Lasso penalty to push the weight sparsity into channel sparsity. In addition, we propose a novel complementary transformed l1 penalty to stabilize the training for extreme compression.
Understanding the 3-dimensional structure of the world is a core challenge in computer vision and robotics. Neural rendering approaches learn an implicit 3D model by predicting what a camera would see from an arbitrary viewpoint. We extend existing neural rendering to more complex, higher dimensional scenes than previously possible. We propose Epipolar Cross Attention (ECA), an attention mechanism that leverages the geometry of the scene to perform efficient non-local operations, requiring only $O(n)$ comparisons per spatial dimension instead of $O(n^2)$. We introduce three new simulated datas…
University of Michigan（密歇根大学）；Google（谷歌）；Google Brain（谷歌大脑）；X Inc（X公司）
This paper focuses on the problem of learning 6-DOF grasping with a parallel jaw gripper in simulation. We propose the notion of a geometry-aware representation in grasping based on the assumption that knowledge of 3D geometry is at the heart of interaction. Our key idea is constraining and regularizing grasping interaction learning through 3D geometry prediction. Specifically, we formulate the learning of deep geometry-aware grasping model in two steps: First, we learn to build mental geometry-aware representation by reconstructing the scene (i.e., 3D occupancy grid) from RGBD input via gener…
University of California, Los Angeles（加利福尼亚大学洛杉矶分校）
Existing pruning techniques preserve deep neural networks' overall ability to make correct predictions but may also amplify hidden biases during the compression process. We propose a novel pruning method, Fairness-aware GRAdient Pruning mEthod (FairGRAPE), that minimizes the disproportionate impacts of pruning on different sub-groups. Our method calculates the per-group importance of each model weight and selects a subset of weights that maintain the relative between-group total importance in pruning. The proposed method then prunes network edges with small importance values and repeats the pr…
Recent adversarial learning research has achieved very impressive progress for modelling cross-domain data shifts in appearance space but its counterpart in modelling cross-domain shifts in geometry space lags far behind. This paper presents an innovative Geometry-Aware Domain Adaptation Network (GA-DAN) that is capable of modelling cross-domain shifts concurrently in both geometry space and appearance space and realistically converting images across domains with very different characteristics. In the proposed GA-DAN, a novel multi-modal spatial learning technique is designed which converts a …
Recently, learning-based ego-motion estimation approaches have drawn strong interest from studies mostly focusing on visual perception. These groundbreaking works focus on unsupervised learning for odometry estimation but mostly for visual sensors. Compared to images, a learning-based approach using Light Detection and Ranging (LiDAR) has been reported in a few studies where, most often, a supervised learning framework is proposed. In this paper, we propose a novel approach to geometry-aware deep LiDAR odometry trainable via both supervised and unsupervised frameworks. We incorporate the Itera…
DP-aware AdaLN-Zero: Taming Conditioning-Induced Heavy-Tailed Gradients in Differentially Private Diffusion
School of Computer Science（计算机科学学院）；Big Data, Minjiang University, Fuzhou, Fujian, China.（大数据，闽江大学，福州，福建，中国）；School of Information, Renmin University of China, Beijing, China（信息学院，中国人民大学，北京，中国）
For responsible decision making in safety-critical settings, machine learning models must effectively detect and process edge-case data. Although existing works show that predictive uncertainty is useful for these tasks, it is not evident from literature which uncertainty-aware models are best suited for a given dataset. Thus, we compare six uncertainty-aware deep learning models on a set of edge-case tasks: robustness to adversarial attacks as well as out-of-distribution and adversarial detection. We find that the geometry of the data sub-manifold is an important factor in determining the suc…
Recently, Sharpness-Aware Minimization (SAM) algorithm has shown state-of-the-art generalization abilities in vision tasks. It demonstrates that flat minima tend to imply better generalization abilities. However, it has some difficulty implying SAM to some natural language tasks, especially to models with drastic gradient changes, such as RNNs. In this work, we analyze the relation between the flatness of the local minimum and its generalization ability from a novel and straightforward theoretical perspective. We propose that the shift of the training and test distributions can be equivalently…
Traditional model-based reinforcement learning approaches learn a model of the environment dynamics without explicitly considering how it will be used by the agent. In the presence of misspecified model classes, this can lead to poor estimates, as some relevant available information is ignored. In this paper, we introduce a novel model-based policy search approach that exploits the knowledge of the current agent policy to learn an approximate transition model, focusing on the portions of the environment that are most relevant for policy improvement. We leverage a weighting scheme, derived from…
First-ish Order Methods: Hessian-aware Scalings of Gradient Descent
School of Mathematics and Physics（数学与物理学院）；University of Queensland（昆士兰大学）；ARC Training Centre for Information Resilience（信息韧性培训中心）；Computer Sciences Department（计算机科学系）；University of Wisconsin-Madison（威斯康星大学麦迪逊分校）
University of Oxford（牛津大学）；Heriot Watt University（赫瑞-瓦特大学）
Inspired by the cognitive process of humans and animals, Curriculum Learning (CL) trains a model by gradually increasing the difficulty of the training data. In this paper, we study whether CL can be applied to complex geometry problems like estimating monocular Visual Odometry (VO). Unlike existing CL approaches, we present a novel CL strategy for learning the geometry of monocular VO by gradually making the learning objective more difficult during training. To this end, we propose a novel geometry-aware objective function by jointly optimizing relative and composite transformations over smal…
SGAT4PASS: Spherical Geometry-Aware Transformer for PAnoramic Semantic Segmentation
College of Computer Science and Technology, Zhejiang University（浙江大学计算机科学与技术学院）；ARC Lab, Tencent PCG（腾讯平台与内容事业群ARC实验室）；Zhejiang University-University of Illinois at Urbana-Champaign Institute, Zhejiang University（浙江大学-伊利诺伊大学厄巴纳-香槟分校研究院）；Zhejiang – Singapore Innovation and AI Joint Research Lab（浙江-新加坡创新与人工智能联合实验室）
Geometry-Aware Deep Congruence Networks for Manifold Learning in Cross-Subject Motor Imagery
Department of Biotechnology（生物技术系）；Indian Institute of Technology Madras（印度理工学院马德拉斯分校）；Department of Data Science and Artificial Intelligence（数据科学与人工智能系）
DAVE: Distribution-aware Attribution via ViT Gradient Decomposition
Jagiellonian University, Faculty of Mathematics and Computer Science（杰洛内大学数学与计算机科学系）；Jagiellonian University, Doctoral School of Exact and Natural Sciences（杰洛内大学精确与自然科学研究博士学院）；Max Planck Institute for Informatics, Saarland Informatics Campus, Saarbrucken, Germany（马克斯·普朗克信息研究所，萨尔布吕肯校区）
Federated learning (FL) is a promising technique that enables many edge devices to train a machine learning model collaboratively in wireless networks. By exploiting the superposition nature of wireless waveforms, over-the-air computation (AirComp) can accelerate model aggregation and hence facilitate communication-efficient FL. Due to channel fading, power control is crucial in AirComp. Prior works assume that the signals to be aggregated from each device, i.e., local gradients have identical statistics. In FL, however, gradient statistics vary over both training iterations and feature dimens…
University of Texas at Austin（德克萨斯大学奥斯汀分校）；Google（谷歌）；Yahoo! Labs（雅虎实验室）
While matrix factorisation models are ubiquitous in large scale recommendation and search, real time application of such models requires inner product computations over an intractably large set of item factors. In this manuscript we present a novel framework that uses the inverted index representation to exploit structural properties of sparse vectors to significantly reduce the run time computational cost of factorisation models. We develop techniques that use geometry aware permutation maps on a tessellated unit sphere to obtain high dimensional sparse embeddings for latent factors with spar…
We propose a novel manifold based geometric approach for learning unsupervised alignment of word embeddings between the source and the target languages. Our approach formulates the alignment learning problem as a domain adaptation problem over the manifold of doubly stochastic matrices. This viewpoint arises from the aim to align the second order information of the two language spaces. The rich geometry of the doubly stochastic manifold allows to employ efficient Riemannian conjugate gradient algorithm for the proposed formulation. Empirically, the proposed approach outperforms state-of-the-ar…
University of Toronto（多伦多大学）；Vector Institute（向量学院）；Max Planck Institute for Intelligent Systems（马克斯·普朗克智能系统研究所）
Standard first-order stochastic optimization algorithms base their updates solely on the average mini-batch gradient, and it has been shown that tracking additional quantities such as the curvature can help de-sensitize common hyperparameters. Based on this intuition, we explore the use of exact per-sample Hessian-vector products and gradients to construct optimizers that are self-tuning and hyperparameter-free. Based on a dynamics model of the gradient, we derive a process which leads to a curvature-corrected, noise-adaptive online gradient estimate. The smoothness of our updates makes it mor…
Score-Aware Policy-Gradient and Performance Guarantees using Local Lyapunov Stability
LAAS–CNRS, Université de Toulouse, CNRS（LAAS–CNRS，图卢兹大学，CNRS）；Eindhoven University of Technology（埃因霍温理工大学）；LAAS-CNRS, IRIT, and Université de Toulouse（LAAS-CNRS，IRIT，图卢兹大学）
In network link prediction, it is possible to hide a target link from being predicted with a small perturbation on network structure. This observation may be exploited in many real world scenarios, for example, to preserve privacy, or to exploit financial security. There have been many recent studies to generate adversarial examples to mislead deep learning models on graph data. However, none of the previous work has considered the dynamic nature of real-world systems. In this work, we present the first study of adversarial attack on dynamic network link prediction (DNLP). The proposed attack …
SAU: Sparsity-Aware Unlearning for LLMs via Gradient Masking and Importance Redistribution
School of Computer Science and Artificial Intelligence, Wuhan University of Technology（武汉理工大学计算机科学与人工智能学院）；University of California San Diego（加州大学圣地亚哥分校）
Equilibrium Selection in Multi-Agent Policy Gradients via Opponent-Aware Basin Entry
Bloomsbury Technology（布洛姆斯伯里技术）；London School of Economics and Political Science（伦敦政治经济学院）；University of Bristol（布里斯托大学）；Johannes Kepler University Linz（林茨约翰尼斯·开普勒大学）；Odesa Polytechnic National University（敖德萨国立技术大学）
