Linfeng Ye

Mingzhuo Li, Guang Li, Linfeng Ye et al.

In this paper, we propose difficulty-guided sampling (DGS) to bridge the target gap between the distillation objective and the downstream task, therefore improving the performance of dataset distillation. Deep neural networks achieve remarkable performance but have time and storage-consuming training processes. Dataset distillation is proposed to generate compact, high-quality distilled datasets, enabling effective model training while maintaining downstream performance. Existing approaches typically focus on features extracted from the original dataset, overlooking task-specific information, which leads to a target gap between the distillation objective and the downstream task. We propose leveraging characteristics that benefit the downstream training into data distillation to bridge this gap. Focusing on the downstream task of image classification, we introduce the concept of difficulty and propose DGS as a plug-in post-stage sampling module. Following the specific target difficulty distribution, the final distilled dataset is sampled from image pools generated by existing methods. We also propose difficulty-aware guidance (DAG) to explore the effect of difficulty in the generation process. Extensive experiments across multiple settings demonstrate the effectiveness of the proposed methods. It also highlights the broader potential of difficulty for diverse downstream tasks.

Shayan Mohajer Hamidi, Xizhen Deng, Renhao Tan et al.

Recently, it was shown that the role of the teacher in knowledge distillation (KD) is to provide the student with an estimate of the true Bayes conditional probability density (BCPD). Notably, the new findings propose that the student's error rate can be upper-bounded by the mean squared error (MSE) between the teacher's output and BCPD. Consequently, to enhance KD efficacy, the teacher should be trained such that its output is close to BCPD in MSE sense. This paper elucidates that training the teacher model with MSE loss equates to minimizing the MSE between its output and BCPD, aligning with its core responsibility of providing the student with a BCPD estimate closely resembling it in MSE terms. In this respect, through a comprehensive set of experiments, we demonstrate that substituting the conventional teacher trained with cross-entropy loss with one trained using MSE loss in state-of-the-art KD methods consistently boosts the student's accuracy, resulting in improvements of up to 2.6\%.

En-Hui Yang, Shayan Mohajer Hamidi, Linfeng Ye et al.

The concepts of conditional mutual information (CMI) and normalized conditional mutual information (NCMI) are introduced to measure the concentration and separation performance of a classification deep neural network (DNN) in the output probability distribution space of the DNN, where CMI and the ratio between CMI and NCMI represent the intra-class concentration and inter-class separation of the DNN, respectively. By using NCMI to evaluate popular DNNs pretrained over ImageNet in the literature, it is shown that their validation accuracies over ImageNet validation data set are more or less inversely proportional to their NCMI values. Based on this observation, the standard deep learning (DL) framework is further modified to minimize the standard cross entropy function subject to an NCMI constraint, yielding CMI constrained deep learning (CMIC-DL). A novel alternating learning algorithm is proposed to solve such a constrained optimization problem. Extensive experiment results show that DNNs trained within CMIC-DL outperform the state-of-the-art models trained within the standard DL and other loss functions in the literature in terms of both accuracy and robustness against adversarial attacks. In addition, visualizing the evolution of learning process through the lens of CMI and NCMI is also advocated.

Houston H. Zhang, Tao Zhang, Baoze Lin et al.

User interface to code (UI2Code) aims to generate executable code that can faithfully reconstruct a given input UI. Prior work focuses largely on web pages and mobile screens, leaving app widgets underexplored. Unlike web or mobile UIs with rich hierarchical context, widgets are compact, context-free micro-interfaces that summarize key information through dense layouts and iconography under strict spatial constraints. Moreover, while (image, code) pairs are widely available for web or mobile UIs, widget designs are proprietary and lack accessible markup. We formalize this setting as the Widget-to-Code (Widget2Code) and introduce an image-only widget benchmark with fine-grained, multi-dimensional evaluation metrics. Benchmarking shows that although generalized multimodal large language models (MLLMs) outperform specialized UI2Code methods, they still produce unreliable and visually inconsistent code. To address these limitations, we develop a baseline that jointly advances perceptual understanding and structured code generation. At the perceptual level, we follow widget design principles to assemble atomic components into complete layouts, equipped with icon retrieval and reusable visualization modules. At the system level, we design an end-to-end infrastructure, WidgetFactory, which includes a framework-agnostic widget-tailored domain-specific language (WidgetDSL) and a compiler that translates it into multiple front-end implementations (e.g., React, HTML/CSS). An adaptive rendering module further refines spatial dimensions to satisfy compactness constraints. Together, these contributions substantially enhance visual fidelity, establishing a strong baseline and unified infrastructure for future Widget2Code research.

Mingzhuo Li, Guang Li, Linfeng Ye et al.

Deep neural networks have achieved impressive performance across a wide range of tasks, but this success often comes with substantial computational and storage costs due to large-scale training data. Dataset distillation addresses this challenge by constructing compact yet informative datasets that enable efficient model training while maintaining downstream performance. However, most existing approaches primarily emphasize matching data distributions or downstream training statistics, with limited attention to preserving high-level semantic information in the distilled data. In this work, we introduce a semantic-aware perspective for dataset distillation by leveraging Contrastive Language-Image Pretraining (CLIP) as a semantic prior for post-sampling. Our goal is to obtain distilled datasets that are not only compact but also semantically class-discriminative and diverse. To this end, we design three semantic scoring functions that quantify class relevance, inter-class separability, and intra-set diversity in a pretrained semantic space. Based on image pools generated by existing distillation methods, we further develop a two-stage strategy for effective sampling: the first stage filters semantically discriminative samples to form a reliable candidate set, and the second stage performs a dynamic diversity-aware selection to reduce redundancy while preserving semantic coverage. Extensive experiments across multiple datasets, image pools, and downstream models demonstrate consistent performance gains, highlighting the effectiveness of incorporating semantic information into dataset distillation.

Linfeng Ye, Zhixiang Chi, Konstantinos N. Plataniotis et al.

In this paper, we propose a novel information theoretic surrogate loss; normalized conditional mutual information (NCMI); as a drop in alternative to the de facto cross-entropy (CE) for training deep neural network (DNN) based classifiers. We first observe that the model's NCMI is inversely proportional to its accuracy. Building on this insight, we introduce an alternating algorithm to efficiently minimize the NCMI. Across image recognition and whole-slide imaging (WSI) subtyping benchmarks, NCMI-trained models surpass state of the art losses by substantial margins at a computational cost comparable to that of CE. Notably, on ImageNet, NCMI yields a 2.77% top-1 accuracy improvement with ResNet-50 comparing to the CE; on CAMELYON-17, replacing CE with NCMI improves the macro-F1 by 8.6% over the strongest baseline. Gains are consistent across various architectures and batch sizes, suggesting that NCMI is a practical and competitive alternative to CE.

Mingzhuo Li, Guang Li, Jiafeng Mao et al.

To alleviate the reliance of deep neural networks on large-scale datasets, dataset distillation aims to generate compact, high-quality synthetic datasets that can achieve comparable performance to the original dataset. The integration of generative models has significantly advanced this field. However, existing approaches primarily focus on aligning the distilled dataset with the original one, often overlooking task-specific information that can be critical for optimal downstream performance. In this paper, focusing on the downstream task of classification, we propose a task-specific sampling strategy for generative dataset distillation that incorporates the concept of difficulty to consider the requirements of the target task better. The final dataset is sampled from a larger image pool with a sampling distribution obtained by matching the difficulty distribution of the original dataset. A logarithmic transformation is applied as a pre-processing step to correct for distributional bias. The results of extensive experiments demonstrate the effectiveness of our method and suggest its potential for enhancing performance on other downstream tasks. The code is available at https://github.com/SumomoTaku/DiffGuideSamp.

Linfeng Ye, Shayan Mohajer Hamidi, Renhao Tan et al.

It is believed that in knowledge distillation (KD), the role of the teacher is to provide an estimate for the unknown Bayes conditional probability distribution (BCPD) to be used in the student training process. Conventionally, this estimate is obtained by training the teacher using maximum log-likelihood (MLL) method. To improve this estimate for KD, in this paper we introduce the concept of conditional mutual information (CMI) into the estimation of BCPD and propose a novel estimator called the maximum CMI (MCMI) method. Specifically, in MCMI estimation, both the log-likelihood and CMI of the teacher are simultaneously maximized when the teacher is trained. Through Eigen-CAM, it is further shown that maximizing the teacher's CMI value allows the teacher to capture more contextual information in an image cluster. Via conducting a thorough set of experiments, we show that by employing a teacher trained via MCMI estimation rather than one trained via MLL estimation in various state-of-the-art KD frameworks, the student's classification accuracy consistently increases, with the gain of up to 3.32\%. This suggests that the teacher's BCPD estimate provided by MCMI method is more accurate than that provided by MLL method. In addition, we show that such improvements in the student's accuracy are more drastic in zero-shot and few-shot settings. Notably, the student's accuracy increases with the gain of up to 5.72\% when 5\% of the training samples are available to the student (few-shot), and increases from 0\% to as high as 84\% for an omitted class (zero-shot). The code is available at \url{https://github.com/iclr2024mcmi/ICLRMCMI}.

Linfeng Ye, Shayan Mohajer Hamidi, En-hui Yang

A deep neural network (DNN) is said to be undistillable if, when used as a black-box input-output teacher, it cannot be distilled through knowledge distillation (KD). In this case, the distilled student (referred to as the knockoff student) does not outperform a student trained independently with label smoothing (LS student) in terms of prediction accuracy. To protect intellectual property of DNNs, it is desirable to build undistillable DNNs. To this end, it is first observed that an undistillable DNN may have the trait that each cluster of its output probability distributions in response to all sample instances with the same label should be highly concentrated to the extent that each cluster corresponding to each label should ideally collapse into one probability distribution. Based on this observation and by measuring the concentration of each cluster in terms of conditional mutual information (CMI), a new training method called CMI minimized (CMIM) method is proposed, which trains a DNN by jointly minimizing the conventional cross entropy (CE) loss and the CMI values of all temperature scaled clusters across the entire temperature spectrum. The resulting CMIM model is shown, by extensive experiments, to be undistillable by all tested KD methods existing in the literature. That is, the knockoff students distilled by these KD methods from the CMIM model underperform the respective LS students. In addition, the CMIM model is also shown to performs better than the model trained with the CE loss alone in terms of their own prediction accuracy.

Shayan Mohajer Hamidi, Linfeng Ye

Federated learning (FL) is a promising technology that enables edge devices/clients to collaboratively and iteratively train a machine learning model under the coordination of a central server. The most common approach to FL is first-order methods, where clients send their local gradients to the server in each iteration. However, these methods often suffer from slow convergence rates. As a remedy, second-order methods, such as quasi-Newton, can be employed in FL to accelerate its convergence. Unfortunately, similarly to the first-order FL methods, the application of second-order methods in FL can lead to unfair models, achieving high average accuracy while performing poorly on certain clients' local datasets. To tackle this issue, in this paper we introduce a novel second-order FL framework, dubbed \textbf{d}istributed \textbf{q}uasi-\textbf{N}ewton \textbf{fed}erated learning (DQN-Fed). This approach seeks to ensure fairness while leveraging the fast convergence properties of quasi-Newton methods in the FL context. Specifically, DQN-Fed helps the server update the global model in such a way that (i) all local loss functions decrease to promote fairness, and (ii) the rate of change in local loss functions aligns with that of the quasi-Newton method. We prove the convergence of DQN-Fed and demonstrate its \textit{linear-quadratic} convergence rate. Moreover, we validate the efficacy of DQN-Fed across a range of federated datasets, showing that it surpasses state-of-the-art fair FL methods in fairness, average accuracy and convergence speed.

Shayan Mohajer Hamidi, Linfeng Ye

Deep neural networks (DNNs) could be deceived by generating human-imperceptible perturbations of clean samples. Therefore, enhancing the robustness of DNNs against adversarial attacks is a crucial task. In this paper, we aim to train robust DNNs by limiting the set of outputs reachable via a norm-bounded perturbation added to a clean sample. We refer to this set as adversarial polytope, and each clean sample has a respective adversarial polytope. Indeed, if the respective polytopes for all the samples are compact such that they do not intersect the decision boundaries of the DNN, then the DNN is robust against adversarial samples. Hence, the inner-working of our algorithm is based on learning \textbf{c}onfined \textbf{a}dversarial \textbf{p}olytopes (CAP). By conducting a thorough set of experiments, we demonstrate the effectiveness of CAP over existing adversarial robustness methods in improving the robustness of models against state-of-the-art attacks including AutoAttack.

Shayan Mohajer Hamidi, Linfeng Ye

Adversarial training (AT) is a popular method for training robust deep neural networks (DNNs) against adversarial attacks. Yet, AT suffers from two shortcomings: (i) the robustness of DNNs trained by AT is highly intertwined with the size of the DNNs, posing challenges in achieving robustness in smaller models; and (ii) the adversarial samples employed during the AT process exhibit poor generalization, leaving DNNs vulnerable to unforeseen attack types. To address these dual challenges, this paper introduces adversarial training via adaptive knowledge amalgamation of an ensemble of teachers (AT-AKA). In particular, we generate a diverse set of adversarial samples as the inputs to an ensemble of teachers; and then, we adaptively amalgamate the logtis of these teachers to train a generalized-robust student. Through comprehensive experiments, we illustrate the superior efficacy of AT-AKA over existing AT methods and adversarial robustness distillation techniques against cutting-edge attacks, including AutoAttack.

Linfeng Ye, Shayan Mohajer Hamidi, Guang Li et al.

Dataset distillation aims to create a compact dataset that retains essential information while maintaining model performance. Diffusion models (DMs) have shown promise for this task but struggle in low images-per-class (IPC) settings, where generated samples lack diversity. In this paper, we address this issue from an information-theoretic perspective by identifying two key types of information that a distilled dataset must preserve: ($i$) prototype information $\mathrm{I}(X;Y)$, which captures label-relevant features; and ($ii$) contextual information $\mathrm{H}(X | Y)$, which preserves intra-class variability. Here, $(X,Y)$ represents the pair of random variables corresponding to the input data and its ground truth label, respectively. Observing that the required contextual information scales with IPC, we propose maximizing $\mathrm{I}(X;Y) + β\mathrm{H}(X | Y)$ during the DM sampling process, where $β$ is IPC-dependent. Since directly computing $\mathrm{I}(X;Y)$ and $\mathrm{H}(X | Y)$ is intractable, we develop variational estimations to tightly lower-bound these quantities via a data-driven approach. Our approach, information-guided diffusion sampling (IGDS), seamlessly integrates with diffusion models and improves dataset distillation across all IPC settings. Experiments on Tiny ImageNet and ImageNet subsets show that IGDS significantly outperforms existing methods, particularly in low-IPC regimes. The code will be released upon acceptance.

Linfeng Ye

Deep Neural Networks (DNNs) have become an integral part of our daily lives, especially in vision-related applications. However, the conventional lossy image compression algorithms are primarily designed for the Human Vision System (HVS), which can non-trivially compromise the DNNs' validation accuracy after compression, as noted in \cite{liu2018deepn}. Thus developing an image compression algorithm for both human and machine (DNNs) is on the horizon. To address the challenge mentioned above, in this paper, we first formulate the image compression as a multi-objective optimization problem which take both human and machine prespectives into account, then we solve it by linear combination, and proposed a novel distortion measure for both human and machine, dubbed Human and Machine-Oriented Error (HMOE). After that, we develop Human And Machine Oriented Soft Decision Quantization (HMOSDQ) based on HMOE, a lossy image compression algorithm for both human and machine (DNNs), and fully complied with JPEG format. In order to evaluate the performance of HMOSDQ, finally we conduct the experiments for two pre-trained well-known DNN-based image classifiers named Alexnet \cite{Alexnet} and VGG-16 \cite{simonyan2014VGG} on two subsets of the ImageNet \cite{deng2009imagenet} validation set: one subset included images with shorter side in the range of 496 to 512, while the other included images with shorter side in the range of 376 to 384. Our results demonstrate that HMOSDQ outperforms the default JPEG algorithm in terms of rate-accuracy and rate-distortion performance. For the Alexnet comparing with the default JPEG algorithm, HMOSDQ can improve the validation accuracy by more than $0.81\%$ at $0.61$ BPP, or equivalently reduce the compression rate of default JPEG by $9.6\times$ while maintaining the same validation accuracy.

Linfeng Ye, Shayan Mohajer Hamidi

The existing work shows that the neural network trained by naive gradient-based optimization method is prone to adversarial attacks, adds small malicious on the ordinary input is enough to make the neural network wrong. At the same time, the attack against a neural network is the key to improving its robustness. The training against adversarial examples can make neural networks resist some kinds of adversarial attacks. At the same time, the adversarial attack against a neural network can also reveal some characteristics of the neural network, a complex high-dimensional non-linear function, as discussed in previous work. In This project, we develop a first-order method to attack the neural network. Compare with other first-order attacks, our method has a much higher success rate. Furthermore, it is much faster than second-order attacks and multi-steps first-order attacks.