CNN
-
An Empirical Study of Handcrafted Feature Learning and Convolutional Neural Networks for Facial Expression Recognition
Abstract
Facial expression recognition is an important computer vision task with applications in human--computer interaction, mental health monitoring, driver alert systems, and behavioral analysis. While convolutional neural networks (CNNs) dominate modern facial expression recognition, handcrafted feature descriptors such as Histogram of Oriented Gradients (HOG) and Local Binary Patterns (LBP) remain useful classical baselines. This study compares HOG with Support Vector Machine (SVM), LBP with Logistic Regression, and a lightweight CNN across three facial expression datasets: FER-2013, CK+, and KDEF. The results show that CNNs achieve the best overall performance, particularly on more complex data, while HOG performs strongly in controlled environments. LBP performs poorly across all datasets. The study highlights that dataset complexity significantly affects performance and that robust feature learning is essential for real-world facial expression recognition.
Abstract (translated)
URL
https://arxiv.org/abs/2607.15288
PDF
-
Interpretable Sperm Morphology Classification via Attention-Guided Deep Learning
2026-06-18 16:19:13Zahra Asghari Varzaneh, Reza Khoshkangini, Thomas Ebner, Lars JohanssonAbstract
Male infertility is a major cause of couple infertility, often linked to abnormal sperm morphology. While deep learning models offer automated analysis, most lack interpretability, limiting their clinical adoption. This study proposes an attention-guided deep learning framework for sperm morphology classification. We combine a pretrained EfficientNet-B0 with a Convolutional Block Attention Module (CBAM) to focus on key areas of the sperm head, improving both accuracy and interpretability. Evaluated on the SMIDS and HuSHem public datasets, our model achieves accuracies of 90.2% and 93.9% (macro F1 scores of 0.913 and 0.948), outperforming SimpleCNN and standard EfficientNet-B0. Furthermore, we use Grad-CAM++ visualizations to highlight features influencing the model's decisions. The results demonstrate that this accurate and transparent framework is a practical tool for automated sperm analysis in fertility clinics.
Abstract (translated)
URL
https://arxiv.org/abs/2606.20438
PDF
-
Leveraging systems' non-linearity to tackle the scarcity of data in the design of Intelligent Fault Diagnosis Systems
2026-06-18 15:02:18Giancarlo Santamato, Andrea Mattia Garavagno, Massimiliano Solazzi, Antonio FrisoliAbstract
Deep Transfer Learning (DTL) allows for the efficient building of Intelligent Fault Diagnosis Systems (IFDS). On the other hand, DTL methods still heavily rely on large amounts of labelled data. Obtaining such an amount of data can be challenging when dealing with machines or structures faults. This document proposes a novel approach to the design of vibration-based IFDS using DTL in condition of strong data scarcity. A periodic multi-excitation level procedure leveraging intrinsic non-linearities of real-world systems is used to produce images that can be conveniently analysed by pre-trained Convolutional Neural Networks (CNNs) to diagnose faults. A new data visualization method and its augmentation technique are proposed in this paper to tackle the typical lack of data encountered during the design of IFDS. Experimental validation on a railway pantograph structure provides effective support for the proposed method.
Abstract (translated)
URL
https://arxiv.org/abs/2606.20323
PDF
-
Hybrid Diffusion Transformer for Instruction-Guided Audio Editing via Rectified Flow
2026-06-18 11:20:08Liting Gao, Yonggang Zhu, Yaru Chen, Dongyu Wang, Shubin Zhang, Zhenbo Li, Jean-Yves Guillemaut, Wenwu WangAbstract
Audio editing aims to modify specific content in an existing audio clip according to a natural language instruction while preserving the remaining acoustic content. Despite the remarkable progress of diffusion models, existing training-based editing methods mainly rely on the local inductive biases and cross-attention interaction in convolutional U-Net backbones, which often hinder long-range semantic alignment and precise understanding and localization of instructions. In contrast, diffusion transformers provide stronger global modeling and multimodal fusion, but existing editing architectures usually adopt a simple stack of MMDiT and DiT blocks. Applying joint attention over concatenated audio and text tokens in all blocks results in quadratic complexity with respect to token length. To balance editing performance and efficiency, we propose a hybrid two-stage diffusion transformer architecture for instruction-guided audio editing based on rectified flow matching. It performs joint attention over audio and text tokens to establish coarse semantic alignment at low-resolution stage, then switches to alternating joint-attention and cross-attention blocks to refine editing details at high-resolution stage. This coarse-to-fine strategy enables efficient and accurate instruction-guided audio editing. Experiments show that the proposed framework achieves notable performance gains on challenging editing tasks involving overlapping audio events and complex instructions, while substantially improving editing efficiency with a compact model.
Abstract (translated)
URL
https://arxiv.org/abs/2606.20101
PDF
-
eCNNTO: A Highly Generalizable ConvNet for Accelerating Topology Optimization
Abstract
This work proposes an element-based Convolutional Neural Network (CNN) to accelerate density-based Topology Optimization (TO), termed eCNNTO. TO generally undergoes a large number of iterations, where finite element analysis is performed in every iteration, leading to the efficiency bottleneck especially when dense meshes are used to achieve high-resolution designs. To address this limitation, eCNNTO is proposed to build upon Kallioras et al. (2020), where a Deep Belief Network (DBN) was trained for every element to predict its near-optimal density from its early history, thereby skipping the great majority of iterations and significantly accelerating the TO procedure. However, the method lacks spatial correlations among neighboring elements and may lead to disconnected features in the final structure. The proposed method employs CNN with residual connections to address this issue. On top of it, a novel training strategy is introduced to further enhance the optimization efficiency, where the training dataset consists of the final stage density histories rather than early ones. This change can also help reduce the required training data size. eCNNTO requires only a small dataset to train and yet it can be generalized to problems with largely different boundary conditions, loading cases, design domain geometries, mesh resolutions, as well as non-design domains. In the end, the generalization capabilities and efficiency of eCNNTO are demonstrated through a variety of examples in two and three dimensions, achieving up to 90% and 97% reduction of iterations, respectively.
Abstract (translated)
URL
https://arxiv.org/abs/2606.19921
PDF
-
Gaussian Process Prior Variational Autoencoder for Endoscopic Videos
2026-06-18 08:03:55Ivan De Boi, Xinxing Shi, Xiaoyu Jiang, Tim J. M. Jaspers, Francisco Caetano, Mauricio A. Alvarez, Fons van der Sommen, Sam Van der JeughtAbstract
Endoscopic video analysis is essential for gastrointestinal diagnosis and computer-assisted interventions, but video sequences are routinely degraded by specular reflections, motion artifacts, and missing frames. These transient corruptions can distract clinicians, reduce image interpretability, and disrupt downstream tasks such as 3D reconstruction and navigation. Effective restoration therefore requires methods that exploit temporal continuity rather than treating frames in isolation. We introduce a Gaussian Process Prior Variational Autoencoder (GPVAE) framework for endoscopic video restoration that replaces the standard factorized latent prior with a temporal Gaussian process prior, enabling interpolation of missing frames with uncertainty-aware reconstruction. The framework combines endoscopy-specific encoders, including a convolutional EndoVAE backbone and pretrained Vision Transformer encoders from GastroNet-5M, with two scalable GP approximations: Hierarchical Prior Approximation (HPA) and Sparse Precision Approximation (SPA). Specular reflections are handled using a DUCKNet-based masking pipeline that excludes corrupted pixels from the reconstruction objective. On the C3VDv2 colonoscopy dataset, the best GPVAE variants reduced image reconstruction RMSE by 21.9\% on average, and by up to 26.1\%, relative to matched VAE baselines. Downstream trajectory RMSE was reduced by 12.7\% on average across classical visual odometry and a pretrained PoseNet, at an average increase of 27.3\% in training time per epoch. Finally, the GP posterior provides per-frame uncertainty estimates that reflect temporal support and offer a confidence signal for restored frames.
Abstract (translated)
URL
https://arxiv.org/abs/2606.19908
PDF
-
SL-S4Wave: Self-Supervised Learning of Physiological Waveforms with Structured State Space Models
2026-06-18 07:46:34Feng Wu, Harsh Deep, Eric Lehman, Sanyam Kapoor, Guoshuai Zhao, Rahul Krishnan, Gari Clifford, Li-wei H LehmanAbstract
Modeling long-sequence medical time series data, such as electrocardiograms (ECG), poses significant challenges due to high sampling rates, multichannel signal complexity, inherent noise, and limited labeled data. While recent self-supervised learning (SSL) methods, based on various encoder architectures such as convolutional neural networks, have been proposed to learn representations from unlabeled data, they often fall short in capturing long-range dependencies and noise-invariant features. Structured state space models (S4) excel at long-sequence modeling, but existing S4 architectures fail to capture the unique characteristics of multichannel physiological waveforms. In this work, we propose SL-S4Wave, a self-supervised learning framework that combines contrastive learning with a tailored encoder built on structured state space models. The encoder incorporates multi-layer global convolution using multiscale subkernels, enabling the capture of both fine-grained local patterns and long-range temporal dependencies in noisy, high-resolution multichannel waveforms. Extensive experiments on real-world datasets demonstrate that SL-S4Wave (1) consistently outperforms state-of-the-art supervised and self-supervised baselines in a challenging arrhythmia detection task, (2) achieves high performance with significantly fewer labeled examples, showcasing strong label efficiency, and (3) maintains robust performance on long waveform segments, highlighting its capacity to model complex temporal dynamics in long sequences that most existing approaches fail to efficiently model, and (4) transfers effectively to unseen arrhythmia types, underscoring its robust cross-domain generalization. We additionally evaluate SL-S4Wave on multiple EEG tasks, achieving superior performance over strong baselines, demonstrating generalizability of our approach beyond cardiac waveforms.
Abstract (translated)
URL
https://arxiv.org/abs/2606.19888
PDF
-
CSWinUNETR: Segmentation of Thin Anatomical Structures in Medical Images
2026-06-18 06:01:31Junho Moon, Haejun Chung, Ikbeom JangAbstract
Accurate segmentation of thin, tortuous anatomical structures, such as retinal vessels, cerebral vasculature, and facial wrinkles, remains challenging due to low contrast, frequent discontinuities, and severe class imbalance. Although recent convolutional and Transformer-based models have improved performance, they often yield fragmented predictions and fail to recover fine branches. We propose CSWinUNETR, a general-purpose backbone for 2D and 3D thin-structure segmentation. It employs cross-shaped stripe self-attention to model long-range principal-axis context and incorporates cyclic shifts to enhance information exchange across stripes. To better preserve fine-grained details, we further introduce a detail-enhanced multi-scale self-attention module that aggregates contextual features from multi-resolution representations. In addition, we propose sparse-control dynamic snake convolution, which reconstructs reliable dense curvilinear kernels from sparsely predicted control points to better follow tortuous geometry. Extensive experiments on four benchmarks across ophthalmology, neurovascular imaging, and dermatology demonstrate that CSWinUNETR consistently outperforms state-of-the-art methods without task-specific post-processing or topology-aware losses. The code is available at this https URL.
Abstract (translated)
URL
https://arxiv.org/abs/2606.19824
PDF
-
GB-LSR: A Fast Local Spectral Image Representation with a Single Global Bandwidth for Continuous Reconstruction and Super-Resolution
2026-06-17 21:50:05Max Shad, Naeem KhoshnevisAbstract
We present GB-LSR (Global-Bandwidth Local Spectral Representation), a fixed-grid local spectral representation for continuous image reconstruction. The image domain is partitioned into non-overlapping square patches, each carrying coefficients for a truncated Fourier basis predicted from shared convolutional-encoder features. A single trainable scalar bandwidth is shared globally across all patches and images, and reconstruction at any continuous coordinate is a fixed-size basis contraction whose cost is independent of image size. We study three bandwidth-handling variants: a trainable global scalar (main), a fixed global scalar, and a per-patch bandwidth field. On a standardized native-reconstruction benchmark across Kodak, Set14, and Urban100, the main variant outperforms matched-budget amortized LIIF / LTE / WIRE re-implementations by 2.8-3.6 dB PSNR and 0.11-0.15 LPIPS, while running at roughly one-quarter of the slowest baseline's inference cost. The single global scalar suffices empirically: per-patch adaptive-bandwidth alternatives do not improve over it on either a closed-form locality diagnostic or an end-to-end ablation. In a separate arbitrary-scale super-resolution (ASR) extension, GB-LSR achieves competitive PSNR-Y under a canonical-style SR protocol and runs 1.44x faster than LIIF-RDN and 3.25x faster than LTE-SwinIR at x4; within the same extension, a variant trained and evaluated without 4-corner local-ensemble averaging gives a 1.77x speedup with 35% lower peak memory and negligible PSNR change, while additionally widening the RDN encoder from 64 to 96 channels gives a small positive PSNR shift with a 1.58x speedup and 31% lower peak memory. Native-reconstruction claims are scoped to the matched-budget amortized protocol, and ASR claims are scoped to a separate canonical-style SR protocol.
Abstract (translated)
URL
https://arxiv.org/abs/2606.19617
PDF
-
ITNet: A Learnable Integral Transform That Subsumes Convolution, Attention, and Recurrence
2026-06-17 19:30:11Ashim Dhor, Rasel Mondal, Pin Yu ChenAbstract
Convolutional networks, recurrent networks, and transformers each encode different inductive biases -- locality, sequential memory, and content-dependent pairwise interaction -- and have remained mathematically distinct since their inception. We show that this fragmentation reflects not a fundamental diversity in how signals should be processed, but rather incomplete views of a single underlying mathematical object: a learnable integral transform. We introduce the Integral Transform Network (ITNet), a unified architecture built around a learnable kernel that depends jointly on positions and features. This kernel is implemented as a small neural network, specifically an MLP, that models pairwise interactions, enabling the model to adapt its behavior from data. We show that convolution, self-attention (including multi-head), and autoregressive recurrence (including LSTM, GRU, S4, and Mamba) arise as special cases under appropriate parameterizations, and that ITNet is a universal approximator of continuous operators. To make this practical, we develop tiled kernel fusion, importance-weighted Monte Carlo integration, and learned low-rank factorization, enabling efficient and scalable computation. A single ITNet architecture with a shared operator and lightweight modality-specific encoders matches or exceeds specialized baselines on ImageNet-1K , GLUE, ModelNet40, VQA\,v2 and NLVR2. The results demonstrate that a single learned interaction mechanism can recover the behavior of all three architectural families from data.
Abstract (translated)
URL
https://arxiv.org/abs/2606.19538
PDF
-
A Distributionally Robust Reinforcement Learning Framework for Constrained Urban EV Dispatch
2026-06-17 14:29:11An Nguyen, Hoang Nguyen, Phuong Le, Hung Pham, Cuong Do, Laurent El GhaouiAbstract
We study city-scale control of electric-vehicle (EV) ride-hailing fleets where dispatch, repositioning, and charging decisions must respect charger and feeder limits under uncertain, spatially correlated demand and travel times. We formulate the problem as a hex-grid semi-Markov decision process (semi-MDP) with mixed actions -- discrete actions for serving, repositioning, and charging, together with continuous charging power -- and variable action durations. To guarantee physical feasibility during both training and deployment, the policy learns over high-level intentions produced by a masked, temperature-annealed actor. These intentions are projected at every decision step through a time-limited rolling mixed-integer linear program (MILP) that strictly enforces state-of-charge, port, and feeder constraints. To mitigate distributional shifts, we optimize a Soft Actor-Critic (SAC) agent against a Wasserstein-1 ambiguity set with a graph-aligned Mahalanobis ground metric that captures spatial correlations. The robust backup uses the Kantorovich-Rubinstein dual, a projected subgradient inner loop, and a primal-dual risk-budget update. Our architecture combines a two-layer Graph Convolutional Network (GCN) encoder, twin critics, and a value network that drives the adversary. Experiments on a large-scale EV fleet simulator built from NYC taxi data show that PD-RSAC achieves the highest net profit, reaching \$1.22M, compared with \$0.58M-\$0.70M for strong heuristic, single-agent RL, and multi-agent RL baselines, including Greedy, SAC, MAPPO, and MADDPG, while maintaining zero feeder-limit violations.
Abstract (translated)
URL
https://arxiv.org/abs/2604.25848
PDF
-
NeuroStrata: An Electroencephalographic Connectivity-Aware Deep Representation Learning Framework for Dynamic Brain Network Analysis of Mental Stress
2026-06-17 07:00:11Sayantan Acharya, Hamzeh Asgharnezhad, Abbas Khosravi, Douglas Creighton, Roohallah Alizadehsani, U Rajendra AcharyaAbstract
This study introduces NeuroStrata, a connectivity-aware deep representation learning framework for EEG-based mental stress analysis using Time-Varying Partial Directed Coherence (TV-PDC). Unlike conventional EEG classification approaches based on static features, NeuroStrata models the temporal evolution of frequency-specific directed connectivity across distributed brain regions. EEG signals from the 32-channel SAM 40 dataset recorded during mental arithmetic tasks were used to generate TV-PDC connectivity maps. These maps were processed using pretrained Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs) to extract deep connectivity embeddings, which were subsequently classified using lightweight machine learning models. Experimental results demonstrate that beta-band connectivity provides the highest discriminative capability, achieving a peak accuracy of 97.3% using the LAION-CLIP-ViT-L14 backbone with a Support Vector Machine classifier, while alpha-band connectivity exhibits consistently stable performance across model configurations. Connectivity analysis revealed prominent frontal-driven alpha influences and centrally integrated beta connectivity patterns associated with stress-related neural dynamics. Temporal evaluation further indicated that classification performance stabilizes in mid-to-late temporal windows, suggesting progressive consolidation of stress-related connectivity signatures. The proposed framework integrates time-varying effective connectivity modelling with deep representation learning to provide an interpretable and automated approach for EEG-based mental stress analysis.
Abstract (translated)
URL
https://arxiv.org/abs/2608.20354
PDF
-
Leveraging Energy Features for Surface Classification with Deep Learning: A Comparative Analysis Across Three Independent Datasets
2026-06-17 05:24:32Alexander Belyaev, Oleg KushnarevAbstract
The energy-based method remains a comparatively underexamined approach for surface classification in mobile robotics, despite promising results in constrained environments. This study evaluated the viability of using energy-derived features as either a standalone classification modality or as supplementary input to inertial data. A comprehensive evaluation was conducted across three publicly available datasets, comparing the performance of modern deep learning architectures including recurrent neural networks, convolutional neural networks, encoder-only transformers, and Mamba state-space models, under automated hyperparameter tuning and input sequence length optimization. The models achieved higher accuracy than previously reported values on all evaluated datasets, with the convolutional neural network yielding the highest overall performance. When relying exclusively on energy-based features, the models attained classification accuracies in the range of 85-90%, approximately 5-10% lower than those achieved when combined with inertial features (96-99%). Augmenting inertial data with energy features resulted in a consistent mean accuracy improvement of 1-2%. These findings indicate that classifiers relying solely on energy features offer sufficient accuracy for standalone deployment, while also providing a consistent gain when used in combination with other sensing modalities.
Abstract (translated)
URL
https://arxiv.org/abs/2606.18698
PDF
-
Real-Time Smart Retail Analytics through Fog Computing using the RetailLightNet Lightweight Model
2026-06-16 19:55:45Mayank Atreya, Navin Chhibber, Shilpa Ajay, R Loshini, K AnithaAbstract
Smart retail systems now depend on the application of real-time video analytics, which can track customer behavior, manage shelf inventory, and enhance in-store processes. Currently cloud-based applications of video analytics face multiple challenges such as high latency of the communication itself, privacy concerns surrounding transmitting raw video, and the computational demands of the heavy vision models used on the often limited network infrastructure. These challenges lead to delays in decision making where prompt action is crucial. To address these short comings, this research proposes Fog-based analytics withRetailLightNet (F RLN), an ultra lightweight deep learning model which is specifically suited for the edge, resource constrained hardware capabilities of devices like the NVIDIA Jetson Nano. The approach utilizes publicly available retail datasets, SKU110K, and RPC, which contains an assortment of product, shelf, and customer interaction scenarios for this http URL processes include preprocessing images through normalization and employing motion-gated frame selection to improve efficiency by omitting redundant frames to ease the workload on the fog processing. Internal within Retail LightNet, image feature extraction is realized through depth wise separable convolutions which help support spatial and channel-wise learning without heavy computational costs. The model employs two prediction heads presence classification, and density regression, supporting shelf monitoring and queue assessment. The proposed F RLN framework attained better results in terms of precision (0.988), recall (0.976), and F1-score(0.976) when compared with existing convolutional neural network. Keywords: fog computing, motion-gated frame selection,NVIDIA jetson, retail lightnet, smart retail.
Abstract (translated)
URL
https://arxiv.org/abs/2610.35871
PDF
-
Deep-Learning-Based Pixelated Microwave Filter Design and Characterization using Electro-Optical Electric-Field Measurements
2026-06-16 18:51:02Han Zhou, Richard Bannister, Caspar Pierce, Haojie Chang, David Widen, Ludvig Fornstedt, Gabriel Melin, Alexander Bohlin, Pontus Lindeberg Fredriksson, Dilbagh Singh, Christian Fager, Koen BuismanAbstract
Traditional microwave filter design typically relies on iterative parameter tuning and predefined topologies, which limits design space and increases development time. This study uses a deep learning approach combining convolutional neural networks with genetic algorithms to automate pixelated microwave filter synthesis. To validate the approach experimentally, both S-parameter and spatial electric-field measurements were analyzed. The synthesized low-pass filter demonstrated excellent agreement between simulated and measured performance, achieving a 7 GHz passband with over 20 dB suppression beyond 9.5 GHz. Electro-optical measurements, for the first time, revealed electric field patterns that resemble coupled transmission-lines or stub structures, providing insight into the emergent characteristics of AI-generated designs.
Abstract (translated)
URL
https://arxiv.org/abs/2606.18402
PDF
-
Deep Learning-Driven Inverse Design of Doherty Power Amplifiers Using Pixelated Combiners and Dual-State Impedance Synthesis
2026-06-16 18:38:15Han Zhou, Haojie Chang, David Widen, Christian FagerAbstract
The output combiner of a Doherty power amplifier (PA) integrates load modulation, impedance matching, and phase compensation within a single network, making its design and synthesis highly challenging. In this paper, we propose a three-port Doherty combiner design methodology that combines deep convolutional neural networks (CNNs), pixelated layout representations, and genetic algorithms (GA) with dual-state impedance synthesis to address both peak and back-off power conditions. As a proof of concept, two GaN HEMT Doherty PA prototypes incorporating three-port pixelated combiners are designed and fabricated. Both prototypes achieve a measured saturated output power exceeding 44.2 dBm with peak drain efficiency above 71.2% within 2.6-2.8 GHz. Furthermore, a drain efficiency as high as 64% is measured at the 6-dB back-off level. After applying digital predistortion, each prototype achieves an adjacent channel leakage ratio (ACLR) better than -51.3 dBc.
Abstract (translated)
URL
https://arxiv.org/abs/2606.18395
PDF
-
Embedded Machine Learning for Microcontroller-Class Edge Devices: Data, Feature, Evaluation, and Deployment Pipelines
2026-06-16 16:22:24Mostafa DarvishiAbstract
Embedded machine learning moves inference from cloud services to resource-constrained devices that must acquire data, preprocess signals, run a model, and act within tight limits on memory, energy, and latency. This paper presents a systems-oriented synthesis of an embedded machine-learning workflow for microcontroller-class platforms. The emphasis is placed on engineering decisions that are often hidden in generic machine-learning introductions: sampling and buffering, feature extraction as dimensionality reduction, validation under class imbalance, model/runtime co-design, and streaming deployment. Two representative signal families are used throughout the paper. The first is inertial motion recognition, where a two-second, three-axis accelerometer window is transformed from raw samples into root-mean-square and spectral features before classification. The second is keyword spotting, where audio is sampled, anti-aliased, transformed into mel-frequency cepstral coefficients, and processed by a compact one-dimensional convolutional network. The paper concludes with practical design rules for robust on-device inference, including data curation, quantization, thresholding, scheduling, and field monitoring.
Abstract (translated)
URL
https://arxiv.org/abs/2606.18122
PDF
-
AIGS-Net: Compact Illumination Field Modeling via 2D Gaussian Splatting for Fast Low-Light Image Enhancement
2026-06-16 14:45:55Yuhan Chen, Kunyang Huang, Fuchen Li, Zhuohan Qin, Guofa Li, Wenbo Chu, Keqiang LiAbstract
Existing low-light image enhancement methods often face a bottleneck between the representation capacity of illumination-field modeling and computational complexity. To address this issue, this paper proposes an Adaptive Illumination Gaussian Splatting Network (AIGS-Net), an ultra-lightweight architecture for fast low-light enhancement. Unlike conventional static priors, AIGS-Net constructs an input-adaptive 2D Gaussian Splatting illumination field. The opacity of Gaussian basis functions is dynamically modulated by relative luminance statistics of the input image, and spatially varying illumination compensation is rendered through ordered alpha compositing. To guide adaptive illumination compensation efficiently, a zero-parameter nonlinear multiscale contextual encoding module is introduced to extract low-frequency structures and local contrast cues without additional convolutional weights. To suppress noise amplification and sensor-induced color bias, AIGS-Net integrates noise-mask estimation, locked single-channel Gamma mapping, cross-channel consistency regularization, and target color-alignment constraints. Experiments on LOL and LSRW benchmarks show that AIGS-Net improves detail recovery and color fidelity while requiring only approximately 40 learnable parameters, achieving an effective trade-off between enhancement quality and extreme inference efficiency.
Abstract (translated)
URL
https://arxiv.org/abs/2606.17998
PDF
-
Functional Equivalence in Attention: A Comprehensive Study with Applications to Linear Mode Connectivity
2026-06-16 11:59:00Viet-Hoang Tran, Vinh Khanh Bui, Van-Hoan Trinh, Tan Lai Ngoc, Tan M. NguyenAbstract
Neural network parameter spaces are inherently non-injective, as distinct parameter configurations can realize identical functions through functional equivalence. While this symmetry is well understood in classical fully connected and convolutional models, it becomes substantially more intricate in modern attention-based architectures. Existing analyses of multihead attention have largely focused on the vanilla formulation, overlooking positional encodings that fundamentally reshape architectural symmetries. In this work, we provide a formal study of functional equivalence in Transformers with positional encodings. Focusing on the two most widely used variants--sinusoidal and rotary positional encodings (RoPE)--we show that sinusoidal encodings preserve the equivalence structure of vanilla attention, whereas rotary encodings significantly reduce the symmetry group, thereby enhancing expressivity. This offers a principled explanation for the growing prominence of RoPE in practice. We further examine how positional encodings affect linear mode connectivity, and through an alignment algorithm, empirically demonstrate that the presence and variability of connectivity across Transformer settings crucially depend on the positional encoding.
Abstract (translated)
URL
https://arxiv.org/abs/2606.17830
PDF
-
A Link between Shock-wave Theory and Symmetry-reduced Stochastic Gradient Descent for Artificial Neural Networks
Abstract
We develop a mathematically explicit link between shock-wave theory and the symmetry-quotiented learning dynamics of stochastic gradient descent, drawing on differential geometry, Lie group theory, and fluid mechanics. Specifically, after quotienting parameter symmetries and applying local-entropy coarse-graining, the effective dynamics satisfy a viscous Hamilton--Jacobi equation on the quotient manifold. Moreover, under the assumption that the raw parameter dynamics can be summarized by a gradient field on the quotiented space, the gradient of the coarse-grained loss function obeys a Burgers-type equation, and shock formation can be established rigorously. We apply our theory to multilayer perceptrons, convolutional neural networks, Transformers, and mean-field networks, and show that they obey the Hamilton--Jacobi or Burgers-type equations. We conjecture that this framework also yields practical diagnostics for deep learning. In architectures such as Transformers, raw parameter norms are often distorted by symmetry redundancy and may therefore be misleading, whereas symmetry-corrected quotient observables provide a principled basis for monitoring, forecasting, and controlling training-phase transitions.
Abstract (translated)
URL
https://arxiv.org/abs/2606.18303
PDF
