Yoojin Choi

Khuong Vo, Mostafa El-Khamy, Yoojin Choi

Photoplethysmography (PPG) is a cost-effective and non-invasive technique that utilizes optical methods to measure cardiac physiology. PPG has become increasingly popular in health monitoring and is used in various commercial and clinical wearable devices. Compared to electrocardiography (ECG), PPG does not provide substantial clinical diagnostic value, despite the strong correlation between the two. Here, we propose a subject-independent attention-based deep state-space model (ADSSM) to translate PPG signals to corresponding ECG waveforms. The model is not only robust to noise but also data-efficient by incorporating probabilistic prior knowledge. To evaluate our approach, 55 subjects' data from the MIMIC-III database were used in their original form, and then modified with noise, mimicking real-world scenarios. Our approach was proven effective as evidenced by the PR-AUC of 0.986 achieved when inputting the translated ECG signals into an existing atrial fibrillation (AFib) detector. ADSSM enables the integration of ECG's extensive knowledge base and PPG's continuous measurement for early diagnosis of cardiovascular disease.

Yoojin Choi, Mostafa El-Khamy, Jungwon Lee

We propose a novel method for training a conditional generative adversarial network (CGAN) without the use of training data, called zero-shot learning of a CGAN (ZS-CGAN). Zero-shot learning of a conditional generator only needs a pre-trained discriminative (classification) model and does not need any training data. In particular, the conditional generator is trained to produce labeled synthetic samples whose characteristics mimic the original training data by using the statistics stored in the batch normalization layers of the pre-trained model. We show the usefulness of ZS-CGAN in data-free quantization of deep neural networks. We achieved the state-of-the-art data-free network quantization of the ResNet and MobileNet classification models trained on the ImageNet dataset. Data-free quantization using ZS-CGAN showed a minimal loss in accuracy compared to that obtained by conventional data-dependent quantization.

Sijia Wang, Yoojin Choi, Junya Chen et al.

Most existing works on continual learning (CL) focus on overcoming the catastrophic forgetting (CF) problem, with dynamic models and replay methods performing exceptionally well. However, since current works tend to assume exclusivity or dissimilarity among learning tasks, these methods require constantly accumulating task-specific knowledge in memory for each task. This results in the eventual prohibitive expansion of the knowledge repository if we consider learning from a long sequence of tasks. In this work, we introduce a paradigm where the continual learner gets a sequence of mixed similar and dissimilar tasks. We propose a new continual learning framework that uses a task similarity detection function that does not require additional learning, with which we analyze whether there is a specific task in the past that is similar to the current task. We can then reuse previous task knowledge to slow down parameter expansion, ensuring that the CL system expands the knowledge repository sublinearly to the number of learned tasks. Our experiments show that the proposed framework performs competitively on widely used computer vision benchmarks such as CIFAR10, CIFAR100, and EMNIST.

Yoojin Choi, Mostafa El-Khamy, Jungwon Lee

This paper proposes two novel knowledge transfer techniques for class-incremental learning (CIL). First, we propose data-free generative replay (DF-GR) to mitigate catastrophic forgetting in CIL by using synthetic samples from a generative model. In the conventional generative replay, the generative model is pre-trained for old data and shared in extra memory for later incremental learning. In our proposed DF-GR, we train a generative model from scratch without using any training data, based on the pre-trained classification model from the past, so we curtail the cost of sharing pre-trained generative models. Second, we introduce dual-teacher information distillation (DT-ID) for knowledge distillation from two teachers to one student. In CIL, we use DT-ID to learn new classes incrementally based on the pre-trained model for old classes and another model (pre-)trained on the new data for new classes. We implemented the proposed schemes on top of one of the state-of-the-art CIL methods and showed the performance improvement on CIFAR-100 and ImageNet datasets.

Yoojin Choi, Jihwan Choi, Mostafa El-Khamy et al.

Network quantization is an essential procedure in deep learning for development of efficient fixed-point inference models on mobile or edge platforms. However, as datasets grow larger and privacy regulations become stricter, data sharing for model compression gets more difficult and restricted. In this paper, we consider data-free network quantization with synthetic data. The synthetic data are generated from a generator, while no data are used in training the generator and in quantization. To this end, we propose data-free adversarial knowledge distillation, which minimizes the maximum distance between the outputs of the teacher and the (quantized) student for any adversarial samples from a generator. To generate adversarial samples similar to the original data, we additionally propose matching statistics from the batch normalization layers for generated data and the original data in the teacher. Furthermore, we show the gain of producing diverse adversarial samples by using multiple generators and multiple students. Our experiments show the state-of-the-art data-free model compression and quantization results for (wide) residual networks and MobileNet on SVHN, CIFAR-10, CIFAR-100, and Tiny-ImageNet datasets. The accuracy losses compared to using the original datasets are shown to be very minimal.

Yoojin Choi, Mostafa El-Khamy, Jungwon Lee

In this paper, we propose a novel variable-rate learned image compression framework with a conditional autoencoder. Previous learning-based image compression methods mostly require training separate networks for different compression rates so they can yield compressed images of varying quality. In contrast, we train and deploy only one variable-rate image compression network implemented with a conditional autoencoder. We provide two rate control parameters, i.e., the Lagrange multiplier and the quantization bin size, which are given as conditioning variables to the network. Coarse rate adaptation to a target is performed by changing the Lagrange multiplier, while the rate can be further fine-tuned by adjusting the bin size used in quantizing the encoded representation. Our experimental results show that the proposed scheme provides a better rate-distortion trade-off than the traditional variable-rate image compression codecs such as JPEG2000 and BPG. Our model also shows comparable and sometimes better performance than the state-of-the-art learned image compression models that deploy multiple networks trained for varying rates.

J. Jon Ryu, Yoojin Choi, Young-Han Kim et al.

A new bimodal generative model is proposed for generating conditional and joint samples, accompanied with a training method with learning a succinct bottleneck representation. The proposed model, dubbed as the variational Wyner model, is designed based on two classical problems in network information theory -- distributed simulation and channel synthesis -- in which Wyner's common information arises as the fundamental limit on the succinctness of the common representation. The model is trained by minimizing the symmetric Kullback--Leibler divergence between variational and model distributions with regularization terms for common information, reconstruction consistency, and latent space matching terms, which is carried out via an adversarial density ratio estimation technique. The utility of the proposed approach is demonstrated through experiments for joint and conditional generation with synthetic and real-world datasets, as well as a challenging zero-shot image retrieval task.

Yoojin Choi, Mostafa El-Khamy, Jungwon Lee

We consider the optimization of deep convolutional neural networks (CNNs) such that they provide good performance while having reduced complexity if deployed on either conventional systems with spatial-domain convolution or lower-complexity systems designed for Winograd convolution. The proposed framework produces one compressed model whose convolutional filters can be made sparse either in the spatial domain or in the Winograd domain. Hence, the compressed model can be deployed universally on any platform, without need for re-training on the deployed platform. To get a better compression ratio, the sparse model is compressed in the spatial domain that has a fewer number of parameters. From our experiments, we obtain $24.2\times$ and $47.7\times$ compressed models for ResNet-18 and AlexNet trained on the ImageNet dataset, while their computational cost is also reduced by $4.5\times$ and $5.1\times$, respectively.

Yoojin Choi, Mostafa El-Khamy, Jungwon Lee

We consider learning deep neural networks (DNNs) that consist of low-precision weights and activations for efficient inference of fixed-point operations. In training low-precision networks, gradient descent in the backward pass is performed with high-precision weights while quantized low-precision weights and activations are used in the forward pass to calculate the loss function for training. Thus, the gradient descent becomes suboptimal, and accuracy loss follows. In order to reduce the mismatch in the forward and backward passes, we utilize mean squared quantization error (MSQE) regularization. In particular, we propose using a learnable regularization coefficient with the MSQE regularizer to reinforce the convergence of high-precision weights to their quantized values. We also investigate how partial L2 regularization can be employed for weight pruning in a similar manner. Finally, combining weight pruning, quantization, and entropy coding, we establish a low-precision DNN compression pipeline. In our experiments, the proposed method yields low-precision MobileNet and ShuffleNet models on ImageNet classification with the state-of-the-art compression ratios of 7.13 and 6.79, respectively. Moreover, we examine our method for image super resolution networks to produce 8-bit low-precision models at negligible performance loss.

Yoojin Choi, Mostafa El-Khamy, Jungwon Lee

We consider the optimization of deep convolutional neural networks (CNNs) such that they provide good performance while having reduced complexity if deployed on either conventional systems utilizing spatial-domain convolution or lower complexity systems designed for Winograd convolution. Furthermore, we explore the universal quantization and compression of these networks. In particular, the proposed framework produces one compressed model whose convolutional filters can be made sparse either in the spatial domain or in the Winograd domain. Hence, one compressed model can be deployed universally on any platform, without need for re-training on the deployed platform, and the sparsity of its convolutional filters can be exploited for further complexity reduction in either domain. To get a better compression ratio, the sparse model is compressed in the spatial domain which has a less number of parameters. From our experiments, we obtain $24.2\times$, $47.7\times$ and $35.4\times$ compressed models for ResNet-18, AlexNet and CT-SRCNN, while their computational cost is also reduced by $4.5\times$, $5.1\times$ and $23.5\times$, respectively.

Yoojin Choi, Mostafa El-Khamy, Jungwon Lee

In this paper, we investigate lossy compression of deep neural networks (DNNs) by weight quantization and lossless source coding for memory-efficient deployment. Whereas the previous work addressed non-universal scalar quantization and entropy coding of DNN weights, we for the first time introduce universal DNN compression by universal vector quantization and universal source coding. In particular, we examine universal randomized lattice quantization of DNNs, which randomizes DNN weights by uniform random dithering before lattice quantization and can perform near-optimally on any source without relying on knowledge of its probability distribution. Moreover, we present a method of fine-tuning vector quantized DNNs to recover the performance loss after quantization. Our experimental results show that the proposed universal DNN compression scheme compresses the 32-layer ResNet (trained on CIFAR-10) and the AlexNet (trained on ImageNet) with compression ratios of $47.1$ and $42.5$, respectively.

Yoojin Choi, Mostafa El-Khamy, Jungwon Lee

Network quantization is one of network compression techniques to reduce the redundancy of deep neural networks. It reduces the number of distinct network parameter values by quantization in order to save the storage for them. In this paper, we design network quantization schemes that minimize the performance loss due to quantization given a compression ratio constraint. We analyze the quantitative relation of quantization errors to the neural network loss function and identify that the Hessian-weighted distortion measure is locally the right objective function for the optimization of network quantization. As a result, Hessian-weighted k-means clustering is proposed for clustering network parameters to quantize. When optimal variable-length binary codes, e.g., Huffman codes, are employed for further compression, we derive that the network quantization problem can be related to the entropy-constrained scalar quantization (ECSQ) problem in information theory and consequently propose two solutions of ECSQ for network quantization, i.e., uniform quantization and an iterative solution similar to Lloyd's algorithm. Finally, using the simple uniform quantization followed by Huffman coding, we show from our experiments that the compression ratios of 51.25, 22.17 and 40.65 are achievable for LeNet, 32-layer ResNet and AlexNet, respectively.