An effective Deep Neural Network (DNN) optimization algorithm that can use decentralized data sets over a peer-to-peer (P2P) network is proposed. In applications such as medical data analysis, the aggregation of data in one location may not be possible due to privacy issues. Hence, we formulate an algorithm to reach a global DNN model that does not require transmission of data among nodes. An existing solution for this issue is gossip stochastic gradient descend (SGD), which updates by averaging node models over a P2P network. However, in practical situations where the data are statistically heterogeneous across the nodes and/or where communication is asynchronous, gossip SGD often gets trapped in local minimum since the model gradients are noticeably different. To overcome this issue, we solve a linearly constrained DNN cost minimization problem, which results in variable update rules that restrict differences among all node models. Our approach can be based on the Primal-Dual Method of Multipliers (PDMM) or the Alternating Direction Method of Multiplier (ADMM), but the cost function is linearized to be suitable for deep learning. It facilitates asynchronous communication. The results of our numerical experiments using CIFAR-10 indicate that the proposed algorithms converge to a global recognition model even though statistically heterogeneous data sets are placed on the nodes.
Rituximab (Rit) chimeric human/mouse monoclonal antibody directed against the CD20 antigen. It can be used before stem cell harvest as a way of in vivo purging or after stem cell transplantation as an adjuvant therapy. Recently, several cases of late-onset neutropenia (LON) have been described in relation to its utilization in stem cell transplantation as well as chemotherapy setting. The reason why Rit induces neutropenia was poorly understood. Dunleavy reported that perturbation of SDF-1 during B-cell recovery retards neutrophil egress from the bone marrow as a mechanism of Rit-induced neutropenia (Blood, Vol. 106, 795-802). We retrospectively analyzed the incidence of LON in autologous peripheral blood stem cell transplantation (autoPBSCT) for malignant lymphoma (ML) using Rit, autoPBSCT for ML without Rit, and autoPBSCT of CD34+ cells for refractory autoimmune disease (AD). Methods LON was defined as neutrophil count less than 1,000/μl after achieving durable hematopoietic engraftment without apparent reasons including myelotoxic drug, viral infection, or disease progression. The hematologic data of out patient clinics were retrospectively analyzed. Results The incidence of LON was 64.3% in autotransplanted ML with Rit. The Nadir of LON was 69-751 (median 578) /μl. The onset of LON from transplant was 61-128 (median 80) days. The duration of LON was 14-124 (median 56) days. There were no LON-related infectious complications. The incidence of LON of autotransplanted ML without Rit and autotransplanted AD was 8.3%, and 10%, respectively. Conclusion When Rit was used peri-autotransplant period, the incidence of LON was increased. Since autotransplantation with purified CD34+ cells did not increase the incidence of LON, our analysis results could not support the idea that perturbation of SDF-1 during B-cell recovery retards neutrophil egress from the bone marrow as a mechanism of Rit-induced neutropenia.Tabled 1ResultsAD using CD34+ cells (n=10)ML with Rit (n=14)ML without Rit (n=12)age(median)14-59(54)31-68(50)21-63 (54)sex (M/F)7 / 710 / 23 / 7diseaseDLBL 12, FL 2DLBL 6, PTCL 2, LBL 1, MCL 1, 2SSc 7, SLE 1, DM/IP 1, WG 1disease status1CR 11, 1Rel 2, 2CR 11CR 8, 1PR 2, Ref 2No. of infused CD34+ cells7.4 (3.9 -13.9) ×10e6/kg8.3 (1.4-24.5) ×10e6/kg4.9 (2.0 -16.0)) ×10e6/kgConditioningR-MCEC 10, MCEC 3, 2MCEC 10, TBI 1, 1HD-CYLON9 (64.3%)1(8.3%)1(10%)Neutoropenia of known reasonVZV 2Drug 1CMV 1observation period (median)121-929 (276)148-1715 (860) Open table in a new tab
We present the task description of the Detection and Classification of Acoustic Scenes and Events (DCASE) 2023 Challenge Task 2: ``First-shot unsupervised anomalous sound detection (ASD) for machine condition monitoring''. The main goal is to enable rapid deployment of ASD systems for new kinds of machines without the need for hyperparameter tuning. In the past ASD tasks, developed methods tuned hyperparameters for each machine type, as the development and evaluation datasets had the same machine types. However, collecting normal and anomalous data as the development dataset can be infeasible in practice. In 2023 Task 2, we focus on solving the first-shot problem, which is the challenge of training a model on a completely novel machine type. Specifically, (i) each machine type has only one section (a subset of machine type) and (ii) machine types in the development and evaluation datasets are completely different. Analysis of 86 submissions from 23 teams revealed that the keys to outperform baselines were: 1) sampling techniques for dealing with class imbalances across different domains and attributes, 2) generation of synthetic samples for robust detection, and 3) use of multiple large pre-trained models to extract meaningful embeddings for the anomaly detector.
The ITU-T Recommendation G.711 is the benchmark standard for narrowband telephony. It has been successful for many decades because of its proven voice quality, ubiquity and utility. A new ITU-T recommendation, denoted G.711.0, has been recently established defining a lossless compression for G.711 packet payloads typically found in IP networks. This paper presents a brief overview of technologies employed within the G.711.0 standard and summarizes the compression and complexity results. It is shown that G.711.0 provides greater than 50% average compression in typical service provider environments while keeping low computational complexity for the encoder/decoder pair (1.0 WMOPS average, <;1.7 WMOPS worst case) and low memory footprint (about 5k octets RAM, 5.7k octets ROM, and 3.6k program memory measured in number of basic operators).
This paper proposes a novel optimization principle and its implementation for unsupervised anomaly detection in sound (ADS) using an autoencoder (AE). The goal of the unsupervised-ADS is to detect unknown anomalous sounds without training data of anomalous sounds. The use of an AE as a normal model is a state-of-the-art technique for the unsupervised-ADS. To decrease the false positive rate (FPR), the AE is trained to minimize the reconstruction error of normal sounds, and the anomaly score is calculated as the reconstruction error of the observed sound. Unfortunately, since this training procedure does not take into account the anomaly score for anomalous sounds, the true positive rate (TPR) does not necessarily increase. In this study, we define an objective function based on the Neyman-Pearson lemma by considering the ADS as a statistical hypothesis test. The proposed objective function trains the AE to maximize the TPR under an arbitrary low FPR condition. To calculate the TPR in the objective function, we consider that the set of anomalous sounds is the complementary set of normal sounds and simulate anomalous sounds by using a rejection sampling algorithm. Through experiments using synthetic data, we found that the proposed method improved the performance measures of the ADS under low FPR conditions. In addition, we confirmed that the proposed method could detect anomalous sounds in real environments.
This paper describes the progress in frequency-domain linear prediction coding (LPC)-based audio coding schemes.Although LPC was originally used only for time-domain speech coders, it has been applied to frequency-domain coders since the late 1980s.With the progress in associated technologies, the frequency-domain LPC-based audio coding scheme has become more promising, and it has been used in speech/audio coding standards, such as MPEG-D unified speech and audio coding and 3GPP enhanced voice services since 2010.Three of the latest investigations on the representations of LPC envelopes in frequency-domain coders are shown.These are the harmonic model, frequency-resolution warping and the Powered All-Pole Spectral Envelope, all of which are aiming at further enhancement of the coding efficiency.
We propose a novel framework for reducing distant noise by using a distributed microphone array; reducing noise propagated from a far distance in real-time. Previous studies have revealed that a distributed microphone array with an instantaneous mixing assumption can effectively reduce noise when the target and noise sources are significantly far apart. However, in distant noise reduction, the target and noise sources are not usually instantaneously mixed because the reverberation-and propagation-time from the noise sources to a microphone is longer than the short-time Fourier transform (STFT) length. To express reverberation- and propagation-parameters, we introduce a multi-delay noise model that represents the reverberation-time as a convolution of the transfer-function-gains and the noise sources and the propagation-time as time-frame delays. These parameters are estimated on the basis of the maximum a posteriori (MAP) estimation. Experimental results show that the proposed method outperformed conventional methods in several performance measurements and could reduce distant noise propagated from more than 100 m away in a real-environment.
