Search

Variable-length feedback coding has the potential to significantly enhance communication reliability in finite block length scenarios by adapting coding strategies based on real-time receiver feedback. Designing such codes, however, is challenging. While deep learning (DL) has been employed to design sophisticated feedback codes, existing DL-aided feedback codes are predominantly fixed-length and suffer performance degradation in the high code rate regime, limiting their adaptability and efficiency. This paper introduces deep variable-length feedback (DeepVLF) code, a novel DL-aided variable-length feedback coding scheme. By segmenting messages into multiple bit groups and employing a threshold-based decoding mechanism for independent decoding of each bit group across successive communication rounds, DeepVLF outperforms existing DL-based feedback codes and establishes a new benchmark in feedback channel coding.

We propose a novel integrated sensing and communication (ISAC) system, where the base station (BS) passively senses the channel parameters using the information carrying signals from a user. To simultaneously guarantee decoding and sensing performance, the user adopts sparse regression codes (SPARCs) with cyclic redundancy check (CRC) to transmit its information bits. The BS generates an initial coarse channel estimation of the parameters after receiving the pilot signal. Then, a novel iterative decoding and parameter sensing algorithm is proposed, where the correctly decoded codewords indicated by the CRC bits are utilized to improve the sensing and channel estimation performance at the BS. In turn, the improved estimate of the channel parameters lead to a better decoding performance. Simulation results show the effectiveness of the proposed iterative decoding and sensing algorithm, where both the sensing and the communication performance are significantly improved with a few iterations. Extensive ablation studies concerning different channel estimation methods and number of CRC bits are carried out for a comprehensive evaluation of the proposed scheme., Comment: 7 pages, conference version

We propose a novel deep learning based method to design a coded waveform for integrated sensing and communication (ISAC) system based on orthogonal frequency-division multiplexing (OFDM). Our ultimate goal is to design a coded waveform, which is capable of providing satisfactory sensing performance of the target while maintaining high communication quality measured in terms of the bit error rate (BER). The proposed LISAC provides an improved waveform design with the assistance of deep neural networks for the encoding and decoding of the information bits. In particular, the transmitter, parameterized by a recurrent neural network (RNN), encodes the input bit sequence into the transmitted waveform for both sensing and communications. The receiver employs a RNN-based decoder to decode the information bits while the transmitter senses the target via maximum likelihood detection. We optimize the system considering both the communication and sensing performance. Simulation results show that the proposed LISAC waveform achieves a better trade-off curve compared to existing alternatives., Comment: 6 pages, conference version

The restless multi-armed bandit (RMAB) framework is a popular model with applications across a wide variety of fields. However, its solution is hindered by the exponentially growing state space (with respect to the number of arms) and the combinatorial action space, making traditional reinforcement learning methods infeasible for large-scale instances. In this paper, we propose GINO-Q, a three-timescale stochastic approximation algorithm designed to learn an asymptotically optimal index policy for RMABs. GINO-Q mitigates the curse of dimensionality by decomposing the RMAB into a series of subproblems, each with the same dimension as a single arm, ensuring that complexity increases linearly with the number of arms. Unlike recently developed Whittle-index-based algorithms, GINO-Q does not require RMABs to be indexable, enhancing its flexibility and applicability. Our experimental results demonstrate that GINO-Q consistently learns near-optimal policies, even for non-indexable RMABs where Whittle-index-based algorithms perform poorly, and it converges significantly faster than existing baselines., Comment: 9 pages, 11 figures

Segmenting multiple objects (e.g., organs) in medical images often requires an understanding of their topology, which simultaneously quantifies the shape of the objects and their positions relative to each other. This understanding is important for segmentation networks to generalize better with limited training data, which is common in medical image analysis. However, many popular networks were trained to optimize only pixel-wise performance, ignoring the topological correctness of the segmentation. In this paper, we introduce a new topology-aware loss function, which we call PI-Att, that explicitly forces the network to minimize the topological dissimilarity between the ground truth and prediction maps. We quantify the topology of each map by the persistence image representation, for the first time in the context of a segmentation network loss. Besides, we propose a new mechanism to adaptively calculate the persistence image at the end of each epoch based on the network's performance. This adaptive calculation enables the network to learn topology outline in the first epochs, and then topology details towards the end of training. The effectiveness of the proposed PI-Att loss is demonstrated on two different datasets for aorta and great vessel segmentation in computed tomography images.

Road traffic congestion prediction is a crucial component of intelligent transportation systems, since it enables proactive traffic management, enhances suburban experience, reduces environmental impact, and improves overall safety and efficiency. Although there are several public datasets, especially for metropolitan areas, these datasets may not be applicable to practical scenarios due to insufficiency in the scale of data (i.e. number of sensors and road links) and several external factors like different characteristics of the target area such as urban, highways and the data collection location. To address this, this paper introduces a novel IBB Traffic graph dataset as an alternative benchmark dataset to mitigate these limitations and enrich the literature with new geographical characteristics. IBB Traffic graph dataset covers the sensor data collected at 2451 distinct locations. Moreover, we propose a novel Road Traffic Prediction Model that strengthens temporal links through feature engineering, node embedding with GLEE to represent inter-related relationships within the traffic network, and traffic prediction with ExtraTrees. The results indicate that the proposed model consistently outperforms the baseline models, demonstrating an average accuracy improvement of 4%.

We consider collaborative inference at the wireless edge, where each client's model is trained independently on their local datasets. Clients are queried in parallel to make an accurate decision collaboratively. In addition to maximizing the inference accuracy, we also want to ensure the privacy of local models. To this end, we leverage the superposition property of the multiple access channel to implement bandwidth-efficient multi-user inference methods. Specifically, we propose different methods for ensemble and multi-view classification that exploit over-the-air computation. We show that these schemes perform better than their orthogonal counterparts with statistically significant differences while using fewer resources and providing privacy guarantees. We also provide experimental results verifying the benefits of the proposed over-the-air multi-user inference approach and perform an ablation study to demonstrate the effectiveness of our design choices. We share the source code of the framework publicly on Github to facilitate further research and reproducibility., Comment: 15 pages, 8 figures. This work extends from our preliminary study presented at the 2022 IEEE International Symposium on Information Theory [1]. arXiv admin note: text overlap with arXiv:2202.03129

Graphs are crucial for representing interrelated data and aiding predictive modeling by capturing complex relationships. Achieving high-quality graph representation is important for identifying linked patterns, leading to improvements in Graph Neural Networks (GNNs) to better capture data structures. However, challenges such as data scarcity, high collection costs, and ethical concerns limit progress. As a result, generative models and data augmentation have become more and more popular. This study explores using generated graphs for data augmentation, comparing the performance of combining generated graphs with real graphs, and examining the effect of different quantities of generated graphs on graph classification tasks. The experiments show that balancing scalability and quality requires different generators based on graph size. Our results introduce a new approach to graph data augmentation, ensuring consistent labels and enhancing classification performance.

Federated learning (FL) has been introduced to enable a large number of clients, possibly mobile devices, to collaborate on generating a generalized machine learning model thanks to utilizing a larger number of local samples without sharing to offer certain privacy to collaborating clients. However, due to the participation of a large number of clients, it is often difficult to profile and verify each client, which leads to a security threat that malicious participants may hamper the accuracy of the trained model by conveying poisoned models during the training. Hence, the aggregation framework at the parameter server also needs to minimize the detrimental effects of these malicious clients. A plethora of attack and defence strategies have been analyzed in the literature. However, often the Byzantine problem is analyzed solely from the outlier detection perspective, being oblivious to the topology of neural networks (NNs). In the scope of this work, we argue that by extracting certain side information specific to the NN topology, one can design stronger attacks. Hence, inspired by the sparse neural networks, we introduce a hybrid sparse Byzantine attack that is composed of two parts: one exhibiting a sparse nature and attacking only certain NN locations with higher sensitivity, and the other being more silent but accumulating over time, where each ideally targets a different type of defence mechanism, and together they form a strong but imperceptible attack. Finally, we show through extensive simulations that the proposed hybrid Byzantine attack is effective against 8 different defence methods.

Recommendation systems can provide accurate recommendations by analyzing user shopping history. A richer user history results in more accurate recommendations. However, in real applications, users prefer e-commerce platforms where the item they seek is at the lowest price. In other words, most users shop from multiple e-commerce platforms simultaneously; different parts of the user's shopping history are shared between different e-commerce platforms. Consequently, we assume in this study that any e-commerce platform has a complete record of the user's history but can only access some parts of it. If a recommendation system is able to predict the missing parts first and enrich the user's shopping history properly, it will be possible to recommend the next item more accurately. Our recommendation system leverages user shopping history to improve prediction accuracy. The proposed approach shows significant improvements in both NDCG@10 and HR@10.

We introduce deep joint source-channel coding (DeepJSCC) schemes for image transmission over cooperative relay channels. The relay either amplifies-and-forwards its received signal, called DeepJSCC-AF, or leverages neural networks to extract relevant features from its received signal, called DeepJSCC-PF (Process-and-Forward). We consider both half- and full-duplex relays, and propose a novel transformer-based model at the relay. For a half-duplex relay, it is shown that the proposed scheme learns to generate correlated signals at the relay and source to obtain beamforming gains. In the full-duplex case, we introduce a novel block-based transmission strategy, in which the source transmits in blocks, and the relay updates its knowledge about the input signal after each block and generates its own signal. To enhance practicality, a single transformer-based model is used at the relay at each block, together with an adaptive transmission module, which allows the model to seamlessly adapt to different channel qualities and the transmission powers}. Simulation results demonstrate the superior performance of DeepJSCC-PF compared to the state-of-the-art BPG image compression algorithm operating at the maximum achievable rate of conventional decode-and-forward and compress-and-forward protocols, in both half- and full-duplex relay scenarios over AWGN and Rayleigh fading channels., Comment: Accepted to IEEE JSAC, 2024

Data availability is crucial for advancing artificial intelligence applications, including voice-based technologies. As content creation, particularly in social media, experiences increasing demand, translation and text-to-speech (TTS) technologies have become essential tools. Notably, the performance of these TTS technologies is highly dependent on the quality of the training data, emphasizing the mutual dependence of data availability and technological progress. This paper introduces an end-to-end tool to generate high-quality datasets for text-to-speech (TTS) models to address this critical need for high-quality data. The contributions of this work are manifold and include: the integration of language-specific phoneme distribution into sample selection, automation of the recording process, automated and human-in-the-loop quality assurance of recordings, and processing of recordings to meet specified formats. The proposed application aims to streamline the dataset creation process for TTS models through these features, thereby facilitating advancements in voice-based technologies., Comment: 9 Pages, 6 Figures, 4 Tables, LREC-COLING 2024

The effectiveness of Intrusion Detection Systems (IDS) is critical in an era where cyber threats are becoming increasingly complex. Machine learning (ML) and deep learning (DL) models provide an efficient and accurate solution for identifying attacks and anomalies in computer networks. However, using ML and DL models in IDS has led to a trust deficit due to their non-transparent decision-making. This transparency gap in IDS research is significant, affecting confidence and accountability. To address, this paper introduces a novel Explainable IDS approach, called X-CBA, that leverages the structural advantages of Graph Neural Networks (GNNs) to effectively process network traffic data, while also adapting a new Explainable AI (XAI) methodology. Unlike most GNN-based IDS that depend on labeled network traffic and node features, thereby overlooking critical packet-level information, our approach leverages a broader range of traffic data through network flows, including edge attributes, to improve detection capabilities and adapt to novel threats. Through empirical testing, we establish that our approach not only achieves high accuracy with 99.47% in threat detection but also advances the field by providing clear, actionable explanations of its analytical outcomes. This research also aims to bridge the current gap and facilitate the broader integration of ML/DL technologies in cybersecurity defenses by offering a local and global explainability solution that is both precise and interpretable.

Despite the crucial importance of addressing Black Hole failures in Internet backbone networks, effective detection strategies in backbone networks are lacking. This is largely because previous research has been centered on Mobile Ad-hoc Networks (MANETs), which operate under entirely different dynamics, protocols, and topologies, making their findings not directly transferable to backbone networks. Furthermore, detecting Black Hole failures in backbone networks is particularly challenging. It requires a comprehensive range of network data due to the wide variety of conditions that need to be considered, making data collection and analysis far from straightforward. Addressing this gap, our study introduces a novel approach for Black Hole detection in backbone networks using specialized Yet Another Next Generation (YANG) data models with Black Hole-sensitive Metric Matrix (BHMM) analysis. This paper details our method of selecting and analyzing four YANG models relevant to Black Hole detection in ISP networks, focusing on routing protocols and ISP-specific configurations. Our BHMM approach derived from these models demonstrates a 10% improvement in detection accuracy and a 13% increase in packet delivery rate, highlighting the efficiency of our approach. Additionally, we evaluate the Machine Learning approach leveraged with BHMM analysis in two different network settings, a commercial ISP network, and a scientific research-only network topology. This evaluation also demonstrates the practical applicability of our method, yielding significantly improved prediction outcomes in both environments.

In this paper, we study the remote estimation problem of a Markov process over a channel with a cost. We formulate this problem as an infinite horizon optimization problem with two players, i.e., a sensor and a monitor, that have distinct information, and with a reward function that takes into account both the communication cost and the estimation quality. We show that the main challenge in solving this problem is associated with the consideration of implicit information, i.e., information that the monitor can obtain about the source when the sensor is silent. Our main objective is to develop a framework for finding solutions to this problem without neglecting implicit information a priori. To that end, we propose three different algorithms. The first one is an alternating policy algorithm that converges to a Nash equilibrium. The second one is an occupancy-state algorithm that is guaranteed to find a globally optimal solution. The last one is a heuristic algorithm that is able to find a near-optimal solution.

This paper introduces a novel approach called "friendly attack" aimed at enhancing the performance of error correction channel codes. Inspired by the concept of adversarial attacks, our method leverages the idea of introducing slight perturbations to the neural network input, resulting in a substantial impact on the network's performance. By introducing small perturbations to fixed-point modulated codewords before transmission, we effectively improve the decoder's performance without violating the input power constraint. The perturbation design is accomplished by a modified iterative fast gradient method. This study investigates various decoder architectures suitable for computing gradients to obtain the desired perturbations. Specifically, we consider belief propagation (BP) for LDPC codes; the error correcting code transformer, BP and neural BP (NBP) for polar codes, and neural BCJR for convolutional codes. We demonstrate that the proposed friendly attack method can improve the reliability across different channels, modulations, codes, and decoders. This method allows us to increase the reliability of communication with a legacy receiver by simply modifying the transmitted codeword appropriately.

Recommendation systems are ubiquitous, from Spotify playlist suggestions to Amazon product suggestions. Nevertheless, depending on the methodology or the dataset, these systems typically fail to capture user preferences and generate general recommendations. Recent advancements in Large Language Models (LLM) offer promising results for analyzing user queries. However, employing these models to capture user preferences and efficiency remains an open question. In this paper, we propose LLMRS, an LLM-based zero-shot recommender system where we employ pre-trained LLM to encode user reviews into a review score and generate user-tailored recommendations. We experimented with LLMRS on a real-world dataset, the Amazon product reviews, for software purchase use cases. The results show that LLMRS outperforms the ranking-based baseline model while successfully capturing meaningful information from product reviews, thereby providing more reliable recommendations.

Acquisition and processing of point clouds (PCs) is a crucial enabler for many emerging applications reliant on 3D spatial data, such as robot navigation, autonomous vehicles, and augmented reality. In most scenarios, PCs acquired by remote sensors must be transmitted to an edge server for fusion, segmentation, or inference. Wireless transmission of PCs not only puts on increased burden on the already congested wireless spectrum, but also confronts a unique set of challenges arising from the irregular and unstructured nature of PCs. In this paper, we meticulously delineate these challenges and offer a comprehensive examination of existing solutions while candidly acknowledging their inherent limitations. In response to these intricacies, we proffer four pragmatic solution frameworks, spanning advanced techniques, hybrid schemes, and distributed data aggregation approaches. In doing so, our goal is to chart a path toward efficient, reliable, and low-latency wireless PC transmission.

We propose a novel hybrid joint source-channel coding (JSCC) scheme for robust image transmission over multi-hop networks. In the considered scenario, a mobile user wants to deliver an image to its destination over a mobile cellular network. We assume a practical setting, where the links between the nodes belonging to the mobile core network are stable and of high quality, while the link between the mobile user and the first node (e.g., the access point) is potentially time-varying with poorer quality. In recent years, neural network based JSCC schemes (called DeepJSCC) have emerged as promising solutions to overcome the limitations of separation-based fully digital schemes. However, relying on analog transmission, DeepJSCC suffers from noise accumulation over multi-hop networks. Moreover, most of the hops within the mobile core network may be high-capacity wireless connections, calling for digital approaches. To this end, we propose a hybrid solution, where DeepJSCC is adopted for the first hop, while the received signal at the first relay is digitally compressed and forwarded through the mobile core network. We show through numerical simulations that the proposed scheme is able to outperform both the fully analog and fully digital schemes. Thanks to DeepJSCC it can avoid the cliff effect over the first hop, while also avoiding noise forwarding over the mobile core network thank to digital transmission. We believe this work paves the way for the practical deployment of DeepJSCC solutions in 6G and future wireless networks., Comment: Accepted to IEEE International Conference on Communications (ICC), 2024. Source code will be released soon

We consider low-latency image transmission over a noisy wireless channel when correlated side information is present only at the receiver side (the Wyner-Ziv scenario). In particular, we are interested in developing practical schemes using a data-driven joint source-channel coding (JSCC) approach, which has been previously shown to outperform conventional separation-based approaches in the practical finite blocklength regimes, and to provide graceful degradation with channel quality. We propose a novel neural network architecture that incorporates the decoder-only side information at multiple stages at the receiver side. Our results demonstrate that the proposed method succeeds in integrating the side information, yielding improved performance at all channel conditions in terms of the various quality measures considered here, especially at low channel signal-to-noise ratios (SNRs) and small bandwidth ratios (BRs). We have made the source code of the proposed method public to enable further research, and the reproducibility of the results., Comment: To appear in IEEE International Conference on Machine Learning for Communication and Networking (ICMLCN) 2024

Under the organization of the base station (BS), wireless federated learning (FL) enables collaborative model training among multiple devices. However, the BS is merely responsible for aggregating local updates during the training process, which incurs a waste of the computational resource at the BS. To tackle this issue, we propose a semi-federated learning (SemiFL) paradigm to leverage the computing capabilities of both the BS and devices for a hybrid implementation of centralized learning (CL) and FL. Specifically, each device sends both local gradients and data samples to the BS for training a shared global model. To improve communication efficiency over the same time-frequency resources, we integrate over-the-air computation for aggregation and non-orthogonal multiple access for transmission by designing a novel transceiver structure. To gain deep insights, we conduct convergence analysis by deriving a closed-form optimality gap for SemiFL and extend the result to two extra cases. In the first case, the BS uses all accumulated data samples to calculate the CL gradient, while a decreasing learning rate is adopted in the second case. Our analytical results capture the destructive effect of wireless communication and show that both FL and CL are special cases of SemiFL. Then, we formulate a non-convex problem to reduce the optimality gap by jointly optimizing the transmit power and receive beamformers. Accordingly, we propose a two-stage algorithm to solve this intractable problem, in which we provide the closed-form solutions to the beamformers. Extensive simulation results on two real-world datasets corroborate our theoretical analysis, and show that the proposed SemiFL outperforms conventional FL and achieves 3.2% accuracy gain on the MNIST dataset compared to state-of-the-art benchmarks., Comment: This paper has been accepted by IEEE Transactions on Wireless Communications

We consider the image transmission problem over a noisy wireless channel via deep learning-based joint source-channel coding (DeepJSCC) along with a denoising diffusion probabilistic model (DDPM) at the receiver. Specifically, we are interested in the perception-distortion trade-off in the practical finite block length regime, in which separate source and channel coding can be highly suboptimal. We introduce a novel scheme, where the conventional DeepJSCC encoder targets transmitting a lower resolution version of the image, which later can be refined thanks to the generative model available at the receiver. In particular, we utilize the range-null space decomposition of the target image; DeepJSCC transmits the range-space of the image, while DDPM progressively refines its null space contents. Through extensive experiments, we demonstrate significant improvements in distortion and perceptual quality of reconstructed images compared to standard DeepJSCC and the state-of-the-art generative learning-based method., Comment: 6 pages, 5 figures. Published at INFOCOM 2024 Workshops

Encoder-decoder networks become a popular choice for various medical image segmentation tasks. When they are trained with a standard loss function, these networks are not explicitly enforced to preserve the shape integrity of an object in an image. However, this ability of the network is important to obtain more accurate results, especially when there is a low-contrast difference between the object and its surroundings. In response to this issue, this work introduces a new shape-aware loss function, which we name FourierLoss. This loss function relies on quantifying the shape dissimilarity between the ground truth and the predicted segmentation maps through the Fourier descriptors calculated on their objects, and penalizing this dissimilarity in network training. Different than the previous studies, FourierLoss offers an adaptive loss function with trainable hyperparameters that control the importance of the level of the shape details that the network is enforced to learn in the training process. This control is achieved by the proposed adaptive loss update mechanism, which end-to-end learns the hyperparameters simultaneously with the network weights by backpropagation. As a result of using this mechanism, the network can dynamically change its attention from learning the general outline of an object to learning the details of its contour points, or vice versa, in different training epochs. Working on 2879 computed tomography images of 93 subjects, our experiments revealed that the proposed adaptive shape-aware loss function led to statistically significantly better results for liver segmentation, compared to its counterparts.

The task of preserving privacy while ensuring efficient communication is a fundamental challenge in federated learning. In this work, we tackle this challenge in the trusted aggregator model, and propose a solution that achieves both objectives simultaneously. We show that employing a quantization scheme based on subtractive dithering at the clients can effectively replicate the normal noise addition process at the aggregator. This implies that we can guarantee the same level of differential privacy against other clients while substantially reducing the amount of communication required, as opposed to transmitting full precision gradients and using central noise addition. We also experimentally demonstrate that the accuracy of our proposed approach matches that of the full precision gradient method.

This work aims to bridge the existing knowledge gap in the optimisation of latency-critical code, specifically focusing on high-frequency trading (HFT) systems. The research culminates in three main contributions: the creation of a Low-Latency Programming Repository, the optimisation of a market-neutral statistical arbitrage pairs trading strategy, and the implementation of the Disruptor pattern in C++. The repository serves as a practical guide and is enriched with rigorous statistical benchmarking, while the trading strategy optimisation led to substantial improvements in speed and profitability. The Disruptor pattern showcased significant performance enhancement over traditional queuing methods. Evaluation metrics include speed, cache utilisation, and statistical significance, among others. Techniques like Cache Warming and Constexpr showed the most significant gains in latency reduction. Future directions involve expanding the repository, testing the optimised trading algorithm in a live trading environment, and integrating the Disruptor pattern with the trading algorithm for comprehensive system benchmarking. The work is oriented towards academics and industry practitioners seeking to improve performance in latency-sensitive applications.

Segmentation networks are not explicitly imposed to learn global invariants of an image, such as the shape of an object and the geometry between multiple objects, when they are trained with a standard loss function. On the other hand, incorporating such invariants into network training may help improve performance for various segmentation tasks when they are the intrinsic characteristics of the objects to be segmented. One example is segmentation of aorta and great vessels in computed tomography (CT) images where vessels are found in a particular geometry in the body due to the human anatomy and they mostly seem as round objects on a 2D CT image. This paper addresses this issue by introducing a new topology-aware loss function that penalizes topology dissimilarities between the ground truth and prediction through persistent homology. Different from the previously suggested segmentation network designs, which apply the threshold filtration on a likelihood function of the prediction map and the Betti numbers of the ground truth, this paper proposes to apply the Vietoris-Rips filtration to obtain persistence diagrams of both ground truth and prediction maps and calculate the dissimilarity with the Wasserstein distance between the corresponding persistence diagrams. The use of this filtration has advantage of modeling shape and geometry at the same time, which may not happen when the threshold filtration is applied. Our experiments on 4327 CT images of 24 subjects reveal that the proposed topology-aware loss function leads to better results than its counterparts, indicating the effectiveness of this use.

The high communication cost of sending model updates from the clients to the server is a significant bottleneck for scalable federated learning (FL). Among existing approaches, state-of-the-art bitrate-accuracy tradeoffs have been achieved using stochastic compression methods -- in which the client $n$ sends a sample from a client-only probability distribution $q_{\phi^{(n)}}$, and the server estimates the mean of the clients' distributions using these samples. However, such methods do not take full advantage of the FL setup where the server, throughout the training process, has side information in the form of a global distribution $p_{\theta}$ that is close to the clients' distribution $q_{\phi^{(n)}}$ in Kullback-Leibler (KL) divergence. In this work, we exploit this closeness between the clients' distributions $q_{\phi^{(n)}}$'s and the side information $p_{\theta}$ at the server, and propose a framework that requires approximately $D_{KL}(q_{\phi^{(n)}}|| p_{\theta})$ bits of communication. We show that our method can be integrated into many existing stochastic compression frameworks to attain the same (and often higher) test accuracy with up to $82$ times smaller bitrate than the prior work -- corresponding to 2,650 times overall compression., Comment: Published at the International Conference on Artificial Intelligence and Statistics (AISTATS), 2024

This paper introduces NoRefER, a novel referenceless quality metric for automatic speech recognition (ASR) systems. Traditional reference-based metrics for evaluating ASR systems require costly ground-truth transcripts. NoRefER overcomes this limitation by fine-tuning a multilingual language model for pair-wise ranking ASR hypotheses using contrastive learning with Siamese network architecture. The self-supervised NoRefER exploits the known quality relationships between hypotheses from multiple compression levels of an ASR for learning to rank intra-sample hypotheses by quality, which is essential for model comparisons. The semi-supervised version also uses a referenced dataset to improve its inter-sample quality ranking, which is crucial for selecting potentially erroneous samples. The results indicate that NoRefER correlates highly with reference-based metrics and their intra-sample ranks, indicating a high potential for referenceless ASR evaluation or a/b testing.

The common standard for quality evaluation of automatic speech recognition (ASR) systems is reference-based metrics such as the Word Error Rate (WER), computed using manual ground-truth transcriptions that are time-consuming and expensive to obtain. This work proposes a multi-language referenceless quality metric, which allows comparing the performance of different ASR models on a speech dataset without ground truth transcriptions. To estimate the quality of ASR hypotheses, a pre-trained language model (LM) is fine-tuned with contrastive learning in a self-supervised learning manner. In experiments conducted on several unseen test datasets consisting of outputs from top commercial ASR engines in various languages, the proposed referenceless metric obtains a much higher correlation with WER scores and their ranks than the perplexity metric from the state-of-art multi-lingual LM in all experiments, and also reduces WER by more than $7\%$ when used for ensembling hypotheses. The fine-tuned model and experiments are made available for the reproducibility: https://github.com/aixplain/NoRefER, Comment: arXiv admin note: substantial text overlap with arXiv:2306.12577

This paper proposes an efficient and semi-automated method for human-in-the-loop post-editing for machine translation (MT) corpus generation. The method is based on online training of a custom MT quality estimation metric on-the-fly as linguists perform post-edits. The online estimator is used to prioritize worse hypotheses for post-editing, and auto-close best hypotheses without post-editing. This way, significant improvements can be achieved in the resulting quality of post-edits at a lower cost due to reduced human involvement. The trained estimator can also provide an online sanity check mechanism for post-edits and remove the need for additional linguists to review them or work on the same hypotheses. In this paper, the effect of prioritizing with the proposed method on the resulting MT corpus quality is presented versus scheduling hypotheses randomly. As demonstrated by experiments, the proposed method improves the lifecycle of MT models by focusing the linguist effort on production samples and hypotheses, which matter most for expanding MT corpora to be used for re-training them.

This paper presents EvolveMT for efficiently combining multiple machine translation (MT) engines. The proposed system selects the output from a single engine for each segment by utilizing online learning techniques to predict the most suitable system for every translation request. A neural quality estimation metric supervises the method without requiring reference translations. The online learning capability of this system allows for dynamic adaptation to alterations in the domain or machine translation engines, thereby obviating the necessity for additional training. EvolveMT selects a subset of translation engines to be called based on the source sentence features. The degree of exploration is configurable according to the desired quality-cost trade-off. Results from custom datasets demonstrate that EvolveMT achieves similar translation accuracy at a lower cost than selecting the best translation of each segment from all translations using an MT quality estimator. To our knowledge, EvolveMT is the first meta MT system that adapts itself after deployment to incoming translation requests from the production environment without needing costly retraining on human feedback.

Federated learning enables distributed training of machine learning (ML) models across multiple devices in a privacy-preserving manner. Hierarchical federated learning (HFL) is further proposed to meet the requirements of both latency and coverage. In this paper, we consider a data-heterogeneous HFL scenario with mobility, mainly targeting vehicular networks. We derive the convergence upper bound of HFL with respect to mobility and data heterogeneity, and analyze how mobility impacts the performance of HFL. While mobility is considered as a challenge from a communication point of view, our goal here is to exploit mobility to improve the learning performance by mitigating data heterogeneity. Simulation results verify the analysis and show that mobility can indeed improve the model accuracy by up to 15.1\% when training a convolutional neural network on the CIFAR-10 dataset using HFL.

3D point cloud is a three-dimensional data format generated by LiDARs and depth sensors, and is being increasingly used in a large variety of applications. This paper presents a novel solution called SEmantic Point cloud Transmission (SEPT), for the transmission of point clouds over wireless channels with limited bandwidth. At the transmitter, SEPT encodes the point cloud via an iterative downsampling and feature extraction process. At the receiver, SEPT reconstructs the point cloud with latent reconstruction and offset-based upsampling. Extensive numerical experiments confirm that SEPT significantly outperforms the standard approach with octree-based compression followed by channel coding. Compared with a more advanced benchmark that utilizes state-of-the-art deep learning-based compression techniques, SEPT achieves comparable performance while eliminating the cliff and leveling effects. Thanks to its improved performance and robustness against channel variations, we believe that SEPT can be instrumental in collaborative sensing and inference applications among robots and vehicles, particularly in the low-latency and high-mobility scenarios., Comment: 7 pages

Data-driven deep learning based code designs, including low-complexity neural decoders for existing codes, or end-to-end trainable auto-encoders have exhibited impressive results, particularly in scenarios for which we do not have high-performing structured code designs. However, the vast majority of existing data-driven solutions for channel coding focus on a point-to-point scenario. In this work, we consider a multiple access channel (MAC) with feedback and try to understand whether deep learning-based designs are capable of enabling coordination and cooperation among the encoders as well as allowing error correction. Simulation results show that the proposed multi-access block attention feedback (MBAF) code improves the upper bound of the achievable rate of MAC without feedback in finite block length regime., Comment: 5 pages