Big Data works perfectly along with Deep learning to extract knowledge from a huge amount of data. However, this processing could take a lot of training time. Genomics is a Big Data science with high dimensionality. It relies on deep learning to solve complicated problems in certain diseases like cancer by using different DNA information such as the transcription factor. TAL1 is a transcription factor that is essential for the development of hematopoiesis and of the vascular system. In this paper, we highlight the potential of deep learning in the field of genomics and its challenges such as the training time that takes hours, weeks, and in some cases months. Therefore, we propose to apply a distributed deep learning implementation based on Convolutional Neural Networks (CNN) that showed good results in decreasing the training time and enhancing the accuracy performance with 95% by using multiple GPU and TPU as accelerators. We proved the efficiency of using a distributed strategy based on data-parallelism in predicting the transcription-factor TAL1 motif faster.
Diffusion models have become a leading method for generative modeling of both image and scientific data. As these models are costly to train and evaluate, reducing the inference cost for diffusion models remains a major goal. Inspired by the recent empirical success in accelerating diffusion models via the parallel sampling technique~\cite{shih2024parallel}, we propose to divide the sampling process into $\mathcal{O}(1)$ blocks with parallelizable Picard iterations within each block. Rigorous theoretical analysis reveals that our algorithm achieves $\widetilde{\mathcal{O}}(\mathrm{poly} \log d)$ overall time complexity, marking the first implementation with provable sub-linear complexity w.r.t. the data dimension $d$. Our analysis is based on a generalized version of Girsanov's theorem and is compatible with both the SDE and probability flow ODE implementations. Our results shed light on the potential of fast and efficient sampling of high-dimensional data on fast-evolving modern large-memory GPU clusters.
Despite advances using low-rank adapters and quantization, pretraining of large models on consumer hardware has not been possible without model sharding, offloading during training, or per-layer gradient updates. To address these limitations, we propose Low-Rank Adapters for Quantized Training (LoQT), a method for efficiently training quantized models. LoQT uses gradient-based tensor factorization to initialize low-rank trainable weight matrices that are periodically merged into quantized full-rank weight matrices. Our approach is suitable for both pretraining and fine-tuning models. We demonstrate this for language modeling and downstream task adaptation, finding that LoQT enables efficient training of models up to 7B parameters on a 24GB GPU. We also demonstrate the feasibility of training a 13B model using per-layer gradient updates on the same hardware.
Despite being a heavily researched topic, Adversarial Training (AT) is rarely, if ever, deployed in practical AI systems for two primary reasons: (i) the gained robustness is frequently accompanied by a drop in generalization and (ii) generating adversarial examples (AEs) is computationally prohibitively expensive. To address these limitations, we propose SMAAT, a new AT algorithm that leverages the manifold conjecture, stating that off-manifold AEs lead to better robustness while on-manifold AEs result in better generalization. Specifically, SMAAT aims at generating a higher proportion of off-manifold AEs by perturbing the intermediate deepnet layer with the lowest intrinsic dimension. This systematically results in better scalability compared to classical AT as it reduces the PGD chains length required for generating the AEs. Additionally, our study provides, to the best of our knowledge, the first explanation for the difference in the generalization and robustness trends between vision and language models, ie., AT results in a drop in generalization in vision models whereas, in encoder-based language models, generalization either improves or remains unchanged. We show that vision transformers and decoder-based models tend to have low intrinsic dimensionality in the earlier layers of the network (more off-manifold AEs), while encoder-based models have low intrinsic dimensionality in the later layers. We demonstrate the efficacy of SMAAT; on several tasks, including robustifying (i) sentiment classifiers, (ii) safety filters in decoder-based models, and (iii) retrievers in RAG setups. SMAAT requires only 25-33% of the GPU time compared to standard AT, while significantly improving robustness across all applications and maintaining comparable generalization.
The automated surface defect detection is a fundamental task in industrial production, and the existing saliencybased works overcome the challenging scenes and give promising detection results. However, the cutting-edge efforts often suffer from large parameter size, heavy computational cost, and slow inference speed, which heavily limits the practical applications. To this end, we devise a multi-scale interactive (MI) module, which employs depthwise convolution (DWConv) and pointwise convolution (PWConv) to independently extract and interactively fuse features of different scales, respectively. Particularly, the MI module can provide satisfactory characterization for defect regions with fewer parameters. Embarking on this module, we propose a lightweight Multi-scale Interactive Network (MINet) to conduct real-time salient object detection of strip steel surface defects. Comprehensive experimental results on SD-Saliency-900 dataset, which contains three kinds of strip steel surface defect detection images (i.e., inclusion, patches, and scratches), demonstrate that the proposed MINet presents comparable detection accuracy with the state-of-the-art methods while running at a GPU speed of 721FPS and a CPU speed of 6.3FPS for 368*368 images with only 0.28M parameters. The code is available at https://github.com/Kunye-Shen/MINet.
Monocular visual simultaneous localization and mapping (V-SLAM) is nowadays an irreplaceable tool in mobile robotics and augmented reality, where it performs robustly. However, human colonoscopies pose formidable challenges like occlusions, blur, light changes, lack of texture, deformation, water jets or tool interaction, which result in very frequent tracking losses. ORB-SLAM3, the top performing multiple-map V-SLAM, is unable to recover from them by merging sub-maps or relocalizing the camera, due to the poor performance of its place recognition algorithm based on ORB features and DBoW2 bag-of-words. We present CudaSIFT-SLAM, the first V-SLAM system able to process complete human colonoscopies in real-time. To overcome the limitations of ORB-SLAM3, we use SIFT instead of ORB features and replace the DBoW2 direct index with the more computationally demanding brute-force matching, being able to successfully match images separated in time for relocation and map merging. Real-time performance is achieved thanks to CudaSIFT, a GPU implementation for SIFT extraction and brute-force matching. We benchmark our system in the C3VD phantom colon dataset, and in a full real colonoscopy from the Endomapper dataset, demonstrating the capabilities to merge sub-maps and relocate in them, obtaining significantly longer sub-maps. Our system successfully maps in real-time 88 % of the frames in the C3VD dataset. In a real screening colonoscopy, despite the much higher prevalence of occluded and blurred frames, the mapping coverage is 53 % in carefully explored areas and 38 % in the full sequence, a 70 % improvement over ORB-SLAM3.
Training large-scale models relies on a vast number of computing resources. For example, training the GPT-4 model (1.8 trillion parameters) requires 25000 A100 GPUs . It is a challenge to build a large-scale cluster with one type of GPU-accelerator. Using multiple types of GPU-accelerators to construct a large-scale cluster is an effective way to solve the problem of insufficient homogeneous GPU-accelerators. However, the existing distributed training systems for large-scale models only support homogeneous GPU-accelerators, not support heterogeneous GPU-accelerators. To address the problem, this paper proposes a distributed training system with hybrid parallelism, HETHUB, for large-scale models, which supports heterogeneous cluster, including AMD, Nvidia GPU and other types of GPU-accelerators . It introduces a distributed unified communicator to realize the communication between heterogeneous GPU-accelerators, a distributed performance predictor, and an automatic parallel planner to develop and train models efficiently with heterogeneous GPU-accelerators. Compared to the distributed training system with homogeneous GPU-accelerators, our system can support six combinations of heterogeneous GPU-accelerators. We train the Llama-140B model on a heterogeneous cluster with 768 GPU-accelerators(128 AMD and 640 GPU-accelerator A). The experiment results show that the optimal performance of our system in the heterogeneous cluster has achieved up to 97.49% of the theoretical upper bound performance.
Efficiently supporting long context length is crucial for Transformer models. The quadratic complexity of the self-attention computation plagues traditional Transformers. Sliding window-based static sparse attention mitigates the problem by limiting the attention scope of the input tokens, reducing the theoretical complexity from quadratic to linear. Although the sparsity induced by window attention is highly structured, it does not align perfectly with the microarchitecture of the conventional accelerators, leading to suboptimal implementation. In response, we propose a dataflow-aware FPGA-based accelerator design, SWAT, that efficiently leverages the sparsity to achieve scalable performance for long input. The proposed microarchitecture is based on a design that maximizes data reuse by using a combination of row-wise dataflow, kernel fusion optimization, and an input-stationary design considering the distributed memory and computation resources of FPGA. Consequently, it achieves up to 22$\times$ and 5.7$\times$ improvement in latency and energy efficiency compared to the baseline FPGA-based accelerator and 15$\times$ energy efficiency compared to GPU-based solution.
The rapid proliferation of generative AI has raised questions about the competitiveness of lower-parameter, locally tunable, open-weight models relative to high-parameter, API-guarded, closed-weight models in terms of performance, domain adaptation, cost, and generalization. Centering under-resourced yet risk-intolerant settings in government, research, and healthcare, we see for-profit closed-weight models as incompatible with requirements for transparency, privacy, adaptability, and standards of evidence. Yet the performance penalty in using open-weight models, especially in low-data and low-resource settings, is unclear. We assess the feasibility of using smaller, open-weight models to replace GPT-4-Turbo in zero-shot, few-shot, and fine-tuned regimes, assuming access to only a single, low-cost GPU. We assess value-sensitive issues around bias, privacy, and abstention on three additional tasks relevant to those topics. We find that with relatively low effort, very low absolute monetary cost, and relatively little data for fine-tuning, small open-weight models can achieve competitive performance in domain-adapted tasks without sacrificing generality. We then run experiments considering practical issues in bias, privacy, and hallucination risk, finding that open models offer several benefits over closed models. We intend this work as a case study in understanding the opportunity cost of reproducibility and transparency over for-profit state-of-the-art zero shot performance, finding this cost to be marginal under realistic settings.
One key challenge in backdoor attacks against large foundation models is the resource limits. Backdoor attacks usually require retraining the target model, which is impractical for very large foundation models. Existing backdoor attacks are mainly designed for supervised classifiers or small foundation models (e.g., BERT). None of these attacks has successfully compromised a very large foundation model, such as Llama-3-70B, especially with limited computational resources. In this paper, we propose TrojFM, a novel backdoor attack tailored for very large foundation models. Our primary technical contribution is the development of a novel backdoor injection method. This method forces a backdoored model to generate similar hidden representations for poisoned inputs regardless of their actual semantics. Our approach injects such backdoors by fine-tuning only a very small proportion of model parameters. This enables TrojFM to efficiently launch downstream task-agnostic backdoor attacks against very large foundation models under limited computational resources. Moreover, we optimize the fine-tuning process with our customized QLoRA technique, enabling launching our attack via only~\textit{one A100 GPU}. Furthermore, we design a new trigger injection method to ensure our attack stealthiness. Through extensive experiments, we first demonstrate that TrojFM can launch effective backdoor attacks against widely used large GPT-style models without jeopardizing their normal functionalities (and outperforming existing attacks on BERT-style models). Furthermore, we show that TrojFM is resilient to SOTA defenses and is insensitive to changes in key hyper-parameters. Finally, we conduct a resource analysis to quantify that our method can significantly save computational and memory costs compared to existing backdoor attacks.
Neural Architecture Search (NAS) has been widely adopted to design neural networks for various computer vision tasks. One of its most promising subdomains is differentiable NAS (DNAS), where the optimal architecture is found in a differentiable manner. However, gradient-based methods suffer from the discretization error, which can severely damage the process of obtaining the final architecture. In our work, we first study the risk of discretization error and show how it affects an unregularized supernet. Then, we present that penalizing high entropy, a common technique of architecture regularization, can hinder the supernet's performance. Therefore, to robustify the DNAS framework, we introduce a novel single-stage searching protocol, which is not reliant on decoding a continuous architecture. Our results demonstrate that this approach outperforms other DNAS methods by achieving 75.3% in the searching stage on the Cityscapes validation dataset and attains performance 1.1% higher than the optimal network of DCNAS on the non-dense search space comprising short connections. The entire training process takes only 5.5 GPU days due to the weight reuse, and yields a computationally efficient architecture. Additionally, we propose a new dataset split procedure, which substantially improves results and prevents architecture degeneration in DARTS.
The poor performance of transformers on arithmetic tasks seems to stem in large part from their inability to keep track of the exact position of each digit inside of a large span of digits. We mend this problem by adding an embedding to each digit that encodes its position relative to the start of the number. In addition to the boost these embeddings provide on their own, we show that this fix enables architectural modifications such as input injection and recurrent layers to improve performance even further. With positions resolved, we can study the logical extrapolation ability of transformers. Can they solve arithmetic problems that are larger and more complex than those in their training data? We find that training on only 20 digit numbers with a single GPU for one day, we can reach state-of-the-art performance, achieving up to 99% accuracy on 100 digit addition problems. Finally, we show that these gains in numeracy also unlock improvements on other multi-step reasoning tasks including sorting and multiplication.
A fundamental problem in learning robust and expressive visual representations lies in efficiently estimating the spatial relationships of visual semantics throughout the entire image. In this study, we propose vHeat, a novel vision backbone model that simultaneously achieves both high computational efficiency and global receptive field. The essential idea, inspired by the physical principle of heat conduction, is to conceptualize image patches as heat sources and model the calculation of their correlations as the diffusion of thermal energy. This mechanism is incorporated into deep models through the newly proposed module, the Heat Conduction Operator (HCO), which is physically plausible and can be efficiently implemented using DCT and IDCT operations with a complexity of $\mathcal{O}(N^{1.5})$. Extensive experiments demonstrate that vHeat surpasses Vision Transformers (ViTs) across various vision tasks, while also providing higher inference speeds, reduced FLOPs, and lower GPU memory usage for high-resolution images. The code will be released at https://github.com/MzeroMiko/vHeat.
LoRA and its variants have become popular parameter-efficient fine-tuning (PEFT) methods due to their ability to avoid excessive computational costs. However, an accuracy gap often exists between PEFT methods and full fine-tuning (FT), and this gap has yet to be systematically studied. In this work, we introduce a method for selecting sparse sub-matrices that aim to minimize the performance gap between PEFT vs. full fine-tuning (FT) while also reducing both fine-tuning computational cost and memory cost. Our Sparse Matrix Tuning (SMT) method begins by identifying the most significant sub-matrices in the gradient update, updating only these blocks during the fine-tuning process. In our experiments, we demonstrate that SMT consistently surpasses other PEFT baseline (e.g. LoRA and DoRA) in fine-tuning popular large language models such as LLaMA across a broad spectrum of tasks, while reducing the GPU memory footprint by 67% compared to FT. We also examine how the performance of LoRA and DoRA tends to plateau and decline as the number of trainable parameters increases, in contrast, our SMT method does not suffer from such issue.
This paper benchmarks and improves existing GPU matrix multiplication algorithms specialized for Kronecker-sparse matrices, whose sparsity patterns are described by Kronecker products. These matrices have recently gained popularity as replacements for dense matrices in neural networks because they preserve accuracy while using fewer parameters. We present the first energy and time benchmarks for the multiplication with such matrices, helping users identify scenarios where Kronecker-sparse matrices are more time- and energy-efficient than their dense counterparts. Our benchmark also reveals that specialized implementations spend up to 50% of their total runtime on memory rewriting operations. To address the challenge of reducing memory transfers, we introduce a new so-called tiling strategy adapted to the Kronecker-sparsity structure, which reduces reads and writes between levels of GPU memory. We implement this tiling strategy in a new CUDA kernel that achieves a median speed-up of x1.4, while also cutting energy consumption by 15%. We further demonstrate the broader impact of our results by applying the new kernel to accelerate transformer inference.
Multi-dimensional parameter spaces are commonly encountered in physics theories that go beyond the Standard Model. However, they often possess complicated posterior geometries that are expensive to traverse using techniques traditional to astroparticle physics. Several recent innovations, which are only beginning to make their way into this field, have made navigating such complex posteriors possible. These include GPU acceleration, automatic differentiation, and neural-network-guided reparameterization. We apply these advancements to dark matter direct detection experiments in the context of non-standard neutrino interactions and benchmark their performances against traditional nested sampling techniques when conducting Bayesian inference. Compared to nested sampling alone, we find that these techniques increase performance for both nested sampling and Hamiltonian Monte Carlo, accelerating inference by factors of $\sim 100$ and $\sim 60$, respectively. As nested sampling also evaluates the Bayesian evidence, these advancements can be exploited to improve model comparison performance while retaining compatibility with existing implementations that are widely used in the natural sciences. Using these techniques, we perform the first scan in the neutrino non-standard interactions parameter space for direct detection experiments whereby all parameters are allowed to vary simultaneously. We expect that these advancements are broadly applicable to other areas of astroparticle physics featuring multi-dimensional parameter spaces.
Federated learning works by aggregating locally computed gradients from multiple clients, thus enabling collaborative training without sharing private client data. However, prior work has shown that the data can actually be recovered by the server using so-called gradient inversion attacks. While these attacks perform well when applied on images, they are limited in the text domain and only permit approximate reconstruction of small batches and short input sequences. In this work, we propose DAGER, the first algorithm to recover whole batches of input text exactly. DAGER leverages the low-rank structure of self-attention layer gradients and the discrete nature of token embeddings to efficiently check if a given token sequence is part of the client data. We use this check to exactly recover full batches in the honest-but-curious setting without any prior on the data for both encoder- and decoder-based architectures using exhaustive heuristic search and a greedy approach, respectively. We provide an efficient GPU implementation of DAGER and show experimentally that it recovers full batches of size up to 128 on large language models (LLMs), beating prior attacks in speed (20x at same batch size), scalability (10x larger batches), and reconstruction quality (ROUGE-1/2 > 0.99).
This paper presents a new self-supervised video representation learning framework, ARVideo, which autoregressively predicts the next video token in a tailored sequence order. Two key designs are included. First, we organize autoregressive video tokens into clusters that span both spatially and temporally, thereby enabling a richer aggregation of contextual information compared to the standard spatial-only or temporal-only clusters. Second, we adopt a randomized spatiotemporal prediction order to facilitate learning from multi-dimensional data, addressing the limitations of a handcrafted spatial-first or temporal-first sequence order. Extensive experiments establish ARVideo as an effective paradigm for self-supervised video representation learning. For example, when trained with the ViT-B backbone, ARVideo competitively attains 81.2% on Kinetics-400 and 70.9% on Something-Something V2, which are on par with the strong benchmark set by VideoMAE. Importantly, ARVideo also demonstrates higher training efficiency, i.e., it trains 14% faster and requires 58% less GPU memory compared to VideoMAE.
This paper introduces StreamV2V, a diffusion model that achieves real-time streaming video-to-video (V2V) translation with user prompts. Unlike prior V2V methods using batches to process limited frames, we opt to process frames in a streaming fashion, to support unlimited frames. At the heart of StreamV2V lies a backward-looking principle that relates the present to the past. This is realized by maintaining a feature bank, which archives information from past frames. For incoming frames, StreamV2V extends self-attention to include banked keys and values and directly fuses similar past features into the output. The feature bank is continually updated by merging stored and new features, making it compact but informative. StreamV2V stands out for its adaptability and efficiency, seamlessly integrating with image diffusion models without fine-tuning. It can run 20 FPS on one A100 GPU, being 15x, 46x, 108x, and 158x faster than FlowVid, CoDeF, Rerender, and TokenFlow, respectively. Quantitative metrics and user studies confirm StreamV2V's exceptional ability to maintain temporal consistency.
In computational digital pathology, accurate nuclear segmentation of Hematoxylin and Eosin (H&E) stained whole slide images (WSIs) is a critical step for many analyses and tissue characterizations. One popular deep learning-based nuclear segmentation approach, HoverNet, offers remarkably accurate results but lacks the high-throughput performance needed for clinical deployment in resource-constrained settings. Our approach, HoverFast, aims to provide fast and accurate nuclear segmentation in H&E images using knowledge distillation from HoverNet. By redesigning the tool with software engineering best practices, HoverFast introduces advanced parallel processing capabilities, efficient data handling, and optimized postprocessing. These improvements facilitate scalable high-throughput performance, making HoverFast more suitable for real-time analysis and application in resource-limited environments. Using a consumer grade Nvidia A5000 GPU, HoverFast showed a 21x speed improvement as compared to HoverNet; reducing mean analysis time for 40x WSIs from ~2 hours to 6 minutes while retaining a concordant mean Dice score of 0.91 against the original HoverNet output. Peak memory usage was also reduced 71% from 44.4GB, to 12.8GB, without requiring SSD-based caching. To ease adoption in research and clinical contexts, HoverFast aligns with best-practices in terms of (a) installation, and (b) containerization, while (c) providing outputs compatible with existing popular open-source image viewing tools such as QuPath. HoverFast has been made open-source and is available at andrewjanowczyk.com/open-source-tools/hoverfast.