Federated Learning (FL) enables collaborative model building among a large number of participants without the need for explicit data sharing. But this approach shows vulnerabilities when privacy inference attacks are applied to it. In particular, in the event of a gradient leakage attack, which has a higher success rate in retrieving sensitive data from the model gradients, FL models are at higher risk due to the presence of communication in their inherent architecture. The most alarming thing about this gradient leakage attack is that it can be performed in such a covert way that it does not hamper the training performance while the attackers backtrack from the gradients to get information about the raw data. Two of the most common approaches proposed as solutions to this issue are homomorphic encryption and adding noise with differential privacy parameters. These two approaches suffer from two major drawbacks. They are: the key generation process becomes tedious with the increasing number of clients, and noise-based differential privacy suffers from a significant drop in global model accuracy. As a countermeasure, we propose a mixed-precision quantized FL scheme, and we empirically show that both of the issues addressed above can be resolved. In addition, our approach can ensure more robustness as different layers of the deep model are quantized with different precision and quantization modes. We empirically proved the validity of our method with three benchmark datasets and found a minimal accuracy drop in the global model after applying quantization.
In the era of billion-parameter-sized Language Models (LMs), start-ups have to follow trends and adapt their technology accordingly. Nonetheless, there are open challenges since the development and deployment of large models comes with a need for high computational resources and has economical consequences. In this work, we follow the steps of the R&D group of a modern legal-tech start-up and present important insights on model development and deployment. We start from ground zero by pre-training multiple domain-specific multi-lingual LMs which are a better fit to contractual and regulatory text compared to the available alternatives (XLM-R). We present benchmark results of such models in a half-public half-private legal benchmark comprising 5 downstream tasks showing the impact of larger model size. Lastly, we examine the impact of a full-scale pipeline for model compression which includes: a) Parameter Pruning, b) Knowledge Distillation, and c) Quantization: The resulting models are much more efficient without sacrificing performance at large.
We introduce a new Langevin dynamics based algorithm, called e-TH$\varepsilon$O POULA, to solve optimization problems with discontinuous stochastic gradients which naturally appear in real-world applications such as quantile estimation, vector quantization, CVaR minimization, and regularized optimization problems involving ReLU neural networks. We demonstrate both theoretically and numerically the applicability of the e-TH$\varepsilon$O POULA algorithm. More precisely, under the conditions that the stochastic gradient is locally Lipschitz in average and satisfies a certain convexity at infinity condition, we establish non-asymptotic error bounds for e-TH$\varepsilon$O POULA in Wasserstein distances and provide a non-asymptotic estimate for the expected excess risk, which can be controlled to be arbitrarily small. Three key applications in finance and insurance are provided, namely, multi-period portfolio optimization, transfer learning in multi-period portfolio optimization, and insurance claim prediction, which involve neural networks with (Leaky)-ReLU activation functions. Numerical experiments conducted using real-world datasets illustrate the superior empirical performance of e-TH$\varepsilon$O POULA compared to SGLD, TUSLA, ADAM, and AMSGrad in terms of model accuracy.
Image copy detection and retrieval from large databases leverage two components. First, a neural network maps an image to a vector representation, that is relatively robust to various transformations of the image. Second, an efficient but approximate similarity search algorithm trades scalability (size and speed) against quality of the search, thereby introducing a source of error. This paper improves the robustness of image copy detection with active indexing, that optimizes the interplay of these two components. We reduce the quantization loss of a given image representation by making imperceptible changes to the image before its release. The loss is back-propagated through the deep neural network back to the image, under perceptual constraints. These modifications make the image more retrievable. Our experiments show that the retrieval and copy detection of activated images is significantly improved. For instance, activation improves by $+40\%$ the Recall1@1 on various image transformations, and for several popular indexing structures based on product quantization and locality sensitivity hashing.
We address the problem of action-conditioned generation of human motion sequences. Existing work falls into two categories: forecast models conditioned on observed past motions, or generative models conditioned on action labels and duration only. In contrast, we generate motion conditioned on observations of arbitrary length, including none. To solve this generalized problem, we propose PoseGPT, an auto-regressive transformer-based approach which internally compresses human motion into quantized latent sequences. An auto-encoder first maps human motion to latent index sequences in a discrete space, and vice-versa. Inspired by the Generative Pretrained Transformer (GPT), we propose to train a GPT-like model for next-index prediction in that space; this allows PoseGPT to output distributions on possible futures, with or without conditioning on past motion. The discrete and compressed nature of the latent space allows the GPT-like model to focus on long-range signal, as it removes low-level redundancy in the input signal. Predicting discrete indices also alleviates the common pitfall of predicting averaged poses, a typical failure case when regressing continuous values, as the average of discrete targets is not a target itself. Our experimental results show that our proposed approach achieves state-of-the-art results on HumanAct12, a standard but small scale dataset, as well as on BABEL, a recent large scale MoCap dataset, and on GRAB, a human-object interactions dataset.
End-to-end Speech Translation (ST) aims at translating the source language speech into target language text without generating the intermediate transcriptions. However, the training of end-to-end methods relies on parallel ST data, which are difficult and expensive to obtain. Fortunately, the supervised data for automatic speech recognition (ASR) and machine translation (MT) are usually more accessible, making zero-shot speech translation a potential direction. Existing zero-shot methods fail to align the two modalities of speech and text into a shared semantic space, resulting in much worse performance compared to the supervised ST methods. In order to enable zero-shot ST, we propose a novel Discrete Cross-Modal Alignment (DCMA) method that employs a shared discrete vocabulary space to accommodate and match both modalities of speech and text. Specifically, we introduce a vector quantization module to discretize the continuous representations of speech and text into a finite set of virtual tokens, and use ASR data to map corresponding speech and text to the same virtual token in a shared codebook. This way, source language speech can be embedded in the same semantic space as the source language text, which can be then transformed into target language text with an MT module. Experiments on multiple language pairs demonstrate that our zero-shot ST method significantly improves the SOTA, and even performers on par with the strong supervised ST baselines.
Driven by significant improvements in architectural design and training pipelines, computer vision has recently experienced dramatic progress in terms of accuracy on classic benchmarks such as ImageNet. These highly-accurate models are challenging to deploy, as they appear harder to compress using standard techniques such as pruning. We address this issue by introducing the Correlation Aware Pruner (CAP), a new unstructured pruning framework which significantly pushes the compressibility limits for state-of-the-art architectures. Our method is based on two technical advancements: a new theoretically-justified pruner, which can handle complex weight correlations accurately and efficiently during the pruning process itself, and an efficient finetuning procedure for post-compression recovery. We validate our approach via extensive experiments on several modern vision models such as Vision Transformers (ViT), modern CNNs, and ViT-CNN hybrids, showing for the first time that these can be pruned to high sparsity levels (e.g. $\geq 75$%) with low impact on accuracy ($\leq 1$% relative drop). Our approach is also compatible with structured pruning and quantization, and can lead to practical speedups of 1.5 to 2.4x without accuracy loss. To further showcase CAP's accuracy and scalability, we use it to show for the first time that extremely-accurate large vision models, trained via self-supervised techniques, can also be pruned to moderate sparsities, with negligible accuracy loss.
For on-device automatic speech recognition (ASR), quantization aware training (QAT) is ubiquitous to achieve the trade-off between model predictive performance and efficiency. Among existing QAT methods, one major drawback is that the quantization centroids have to be predetermined and fixed. To overcome this limitation, we introduce a regularization-free, "soft-to-hard" compression mechanism with self-adjustable centroids in a mu-Law constrained space, resulting in a simpler yet more versatile quantization scheme, called General Quantizer (GQ). We apply GQ to ASR tasks using Recurrent Neural Network Transducer (RNN-T) and Conformer architectures on both LibriSpeech and de-identified far-field datasets. Without accuracy degradation, GQ can compress both RNN-T and Conformer into sub-8-bit, and for some RNN-T layers, to 1-bit for fast and accurate inference. We observe a 30.73% memory footprint saving and 31.75% user-perceived latency reduction compared to 8-bit QAT via physical device benchmarking.
Quantizing neural networks to low-bitwidth is important for model deployment on resource-limited edge hardware. Although a quantized network has a smaller model size and memory footprint, it is fragile to adversarial attacks. However, few methods study the robustness and training efficiency of quantized networks. To this end, we propose a new method by recasting robust quantization as an online domain generalization problem, termed ODG-Q, which generates diverse adversarial data at a low cost during training. ODG-Q consistently outperforms existing works against various adversarial attacks. For example, on CIFAR-10 dataset, ODG-Q achieves 49.2% average improvements under five common white-box attacks and 21.7% average improvements under five common black-box attacks, with a training cost similar to that of natural training (viz. without adversaries). To our best knowledge, this work is the first work that trains both quantized and binary neural networks on ImageNet that consistently improve robustness under different attacks. We also provide a theoretical insight of ODG-Q that accounts for the bound of model risk on attacked data.
Deep learning applications are being transferred from the cloud to edge with the rapid development of embedded computing systems. In order to achieve higher energy efficiency with the limited resource budget, neural networks(NNs) must be carefully designed in two steps, the architecture design and the quantization policy choice. Neural Architecture Search(NAS) and Quantization have been proposed separately when deploying NNs onto embedded devices. However, taking the two steps individually is time-consuming and leads to a sub-optimal final deployment. To this end, we propose a novel neural network design framework called Hardware-aware Quantized Neural Architecture Search(HQNAS) framework which combines the NAS and Quantization together in a very efficient manner using weight-sharing and bit-sharing. It takes only 4 GPU hours to discover an outstanding NN policy on CIFAR10. It also takes only %10 GPU time to generate a comparable model on Imagenet compared to the traditional NAS method with 1.8x decrease of latency and a negligible accuracy loss of only 0.7%. Besides, our method can be adapted in a lifelong situation where the neural network needs to evolve occasionally due to changes of local data, environment and user preference.
Due to user privacy and regulatory restrictions, federate learning (FL) is proposed as a distributed learning framework for training deep neural networks (DNN) on decentralized data clients. Recent advancements in FL have applied Neural Architecture Search (NAS) to replace the predefined one-size-fit-all DNN model, which is not optimal for all tasks of various data distributions, with searchable DNN architectures. However, previous methods suffer from expensive communication cost rasied by frequent large model parameters transmission between the server and clients. Such difficulty is further amplified when combining NAS algorithms, which commonly require prohibitive computation and enormous model storage. Towards this end, we propose FAQS, an efficient personalized FL-NAS-Quantization framework to reduce the communication cost with three features: weight-sharing super kernels, bit-sharing quantization and masked transmission. FAQS has an affordable search time and demands very limited size of transmitted messages at each round. By setting different personlized pareto function loss on local clients, FAQS can yield heterogeneous hardware-aware models for various user preferences. Experimental results show that FAQS achieves average reduction of 1.58x in communication bandwith per round compared with normal FL framework and 4.51x compared with FL+NAS framwork.
Model compression is vital to the deployment of deep learning on edge devices. Low precision representations, achieved via quantization of weights and activations, can reduce inference time and memory requirements. However, quantifying and predicting the response of a model to the changes associated with this procedure remains challenging. This response is non-linear and heterogeneous throughout the network. Understanding which groups of parameters and activations are more sensitive to quantization than others is a critical stage in maximizing efficiency. For this purpose, we propose FIT. Motivated by an information geometric perspective, FIT combines the Fisher information with a model of quantization. We find that FIT can estimate the final performance of a network without retraining. FIT effectively fuses contributions from both parameter and activation quantization into a single metric. Additionally, FIT is fast to compute when compared to existing methods, demonstrating favourable convergence properties. These properties are validated experimentally across hundreds of quantization configurations, with a focus on layer-wise mixed-precision quantization.
Deep reinforcement learning (DRL) is one of the most powerful tools for synthesizing complex robotic behaviors. But training DRL models is incredibly compute and memory intensive, requiring large training datasets and replay buffers to achieve performant results. This poses a challenge for the next generation of field robots that will need to learn on the edge to adapt to their environment. In this paper, we begin to address this issue through observation space quantization. We evaluate our approach using four simulated robot locomotion tasks and two state-of-the-art DRL algorithms, the on-policy Proximal Policy Optimization (PPO) and off-policy Soft Actor-Critic (SAC) and find that observation space quantization reduces overall memory costs by as much as 4.2x without impacting learning performance.
Secure aggregation is commonly used in federated learning (FL) to alleviate privacy concerns related to the central aggregator seeing all parameter updates in the clear. Unfortunately, most existing secure aggregation schemes ignore two critical orthogonal research directions that aim to (i) significantly reduce client-server communication and (ii) mitigate the impact of malicious clients. However, both of these additional properties are essential to facilitate cross-device FL with thousands or even millions of (mobile) participants. In this paper, we unite both research directions by introducing ScionFL, the first secure aggregation framework for FL that operates efficiently on quantized inputs and simultaneously provides robustness against malicious clients. Our framework leverages (novel) multi-party computation (MPC) techniques and supports multiple linear (1-bit) quantization schemes, including ones that utilize the randomized Hadamard transform and Kashin's representation. Our theoretical results are supported by extensive evaluations. We show that with no overhead for clients and moderate overhead for the server compared to transferring and processing quantized updates in plaintext, we obtain comparable accuracy for standard FL benchmarks. Moreover, we demonstrate the robustness of our framework against state-of-the-art poisoning attacks.
This paper presents an optimized methodology to design and deploy Speech Enhancement (SE) algorithms based on Recurrent Neural Networks (RNNs) on a state-of-the-art MicroController Unit (MCU), with 1+8 general-purpose RISC-V cores. To achieve low-latency execution, we propose an optimized software pipeline interleaving parallel computation of LSTM or GRU recurrent blocks, featuring vectorized 8-bit integer (INT8) and 16-bit floating-point (FP16) compute units, with manually-managed memory transfers of model parameters. To ensure minimal accuracy degradation with respect to the full-precision models, we propose a novel FP16-INT8 Mixed-Precision Post-Training Quantization (PTQ) scheme that compresses the recurrent layers to 8-bit while the bit precision of remaining layers is kept to FP16. Experiments are conducted on multiple LSTM and GRU based SE models trained on the Valentini dataset, featuring up to 1.24M parameters. Thanks to the proposed approaches, we speed-up the computation by up to 4x with respect to the lossless FP16 baselines. Differently from a uniform 8-bit quantization that degrades the PESQ score by 0.3 on average, the Mixed-Precision PTQ scheme leads to a low-degradation of only 0.06, while achieving a 1.4-1.7x memory saving. Thanks to this compression, we cut the power cost of the external memory by fitting the large models on the limited on-chip non-volatile memory and we gain a MCU power saving of up to 2.5x by reducing the supply voltage from 0.8V to 0.65V while still matching the real-time constraints. Our design results 10x more energy efficient than state-of-the-art SE solutions deployed on single-core MCUs that make use of smaller models and quantization-aware training.
The biggest challenge for the deployment of Deep Neural Networks (DNNs) close to the generated data on edge devices is their size, i.e., memory footprint and computational complexity. Both are significantly reduced with quantization. With the resulting lower word-length, the energy efficiency of DNNs increases proportionally. However, lower word-length typically causes accuracy degradation. To counteract this effect, the quantized DNN is retrained. Unfortunately, training costs up to 5000x more energy than the inference of the quantized DNN. To address this issue, we propose a post-training quantization flow without the need for retraining. For this, we investigated different quantization options. Furthermore, our analysis systematically assesses the impact of reduced word-lengths of weights and activations revealing a clear trend for the choice of word-length. Both aspects have not been systematically investigated so far. Our results are independent of the depth of the DNNs and apply to uniform quantization, allowing fast quantization of a given pre-trained DNN. We excel state-of-the-art for 6 bit by 2.2% Top-1 accuracy for ImageNet. Without retraining, our quantization to 8 bit surpasses floating-point accuracy.
Collaborative Intelligence (CI) has emerged as a promising framework for deploying Artificial Intelligence (AI) models on resource-constrained edge devices. In CI, the AI model is partitioned between the edge device and the cloud, with intermediate features transmitted from the edge sub-model to the cloud sub-model to complete the inference task. However, reducing feature transmission overhead while maintaining task performance remains a challenge, particularly in the case of noisy wireless channels. In this paper, we propose a Spiking Neural Network (SNN)-based Semantic Communication (SC) model, SNN-SC, which extracts compact semantic information from features and transmits it through digital binary channels. Compared to the Deep Neural Network (DNN)-based SC model, whose output is floating-point, the binary output of SNN makes SNN-SC directly applicable to digital binary channels without the need for extra quantization. Moreover, we introduce a novel spiking neuron called IHF to enhance the reconstruction capability of the SNN-SC decoder. Finally, we enhance the performance of SNN-SC by maximizing the entropy of semantic information. SNN-SC achieves a higher compression ratio and overcomes the `cliff effect' compared to the traditional separate source and channel coding method. In addition, SNN-SC has lower computational complexity than the DNN-based SC model and maintains higher task performance under poor channel conditions.
Quantization is an effective technique to reduce memory footprint, inference latency, and power consumption of deep learning models. However, existing quantization methods suffer from accuracy degradation compared to full-precision (FP) models due to the errors introduced by coarse gradient estimation through non-differentiable quantization layers. The existence of sharp local minima in the loss landscapes of overparameterized models (e.g., Transformers) tends to aggravate such performance penalty in low-bit (2, 4 bits) settings. In this work, we propose sharpness- and quantization-aware training (SQuAT), which would encourage the model to converge to flatter minima while performing quantization-aware training. Our proposed method alternates training between sharpness objective and step-size objective, which could potentially let the model learn the most suitable parameter update magnitude to reach convergence near-flat minima. Extensive experiments show that our method can consistently outperform state-of-the-art quantized BERT models under 2, 3, and 4-bit settings on GLUE benchmarks by 1%, and can sometimes even outperform full precision (32-bit) models. Our experiments on empirical measurement of sharpness also suggest that our method would lead to flatter minima compared to other quantization methods.
The large pre-trained vision transformers (ViTs) have demonstrated remarkable performance on various visual tasks, but suffer from expensive computational and memory cost problems when deployed on resource-constrained devices. Among the powerful compression approaches, quantization extremely reduces the computation and memory consumption by low-bit parameters and bit-wise operations. However, low-bit ViTs remain largely unexplored and usually suffer from a significant performance drop compared with the real-valued counterparts. In this work, through extensive empirical analysis, we first identify the bottleneck for severe performance drop comes from the information distortion of the low-bit quantized self-attention map. We then develop an information rectification module (IRM) and a distribution guided distillation (DGD) scheme for fully quantized vision transformers (Q-ViT) to effectively eliminate such distortion, leading to a fully quantized ViTs. We evaluate our methods on popular DeiT and Swin backbones. Extensive experimental results show that our method achieves a much better performance than the prior arts. For example, our Q-ViT can theoretically accelerates the ViT-S by 6.14x and achieves about 80.9% Top-1 accuracy, even surpassing the full-precision counterpart by 1.0% on ImageNet dataset. Our codes and models are attached on https://github.com/YanjingLi0202/Q-ViT
Soft errors in large VLSI circuits pose dramatic influence on computing- and memory-intensive neural network (NN) processing. Understanding the influence of soft errors on NNs is critical to protect against soft errors for reliable NN processing. Prior work mainly rely on fault simulation to analyze the influence of soft errors on NN processing. They are accurate but usually specific to limited configurations of errors and NN models due to the prohibitively slow simulation speed especially for large NN models and datasets. With the observation that the influence of soft errors propagates across a large number of neurons and accumulates as well, we propose to characterize the soft error induced data disturbance on each neuron with normal distribution model according to central limit theorem and develop a series of statistical models to analyze the behavior of NN models under soft errors in general. The statistical models reveal not only the correlation between soft errors and NN model accuracy, but also how NN parameters such as quantization and architecture affect the reliability of NNs. The proposed models are compared with fault simulation and verified comprehensively. In addition, we observe that the statistical models that characterize the soft error influence can also be utilized to predict fault simulation results in many cases and we explore the use of the proposed statistical models to accelerate fault simulations of NNs. According to our experiments, the accelerated fault simulation shows almost two orders of magnitude speedup with negligible simulation accuracy loss over the baseline fault simulations.