Search

The biological brain has inspired multiple advances in machine learning. However, most state-of-the-art models in computer vision do not operate like the human brain, simply because they are not capable of changing or improving their decisions/outputs based on a deeper analysis. The brain is recurrent, while these models are not. It is therefore relevant to explore what would be the impact of adding recurrent mechanisms to existing state-of-the-art architectures and to answer the question of whether recurrency can improve existing architectures. To this end, we build on a feed-forward segmentation model and explore multiple types of recurrency for image segmentation. We explore self-organizing, relational, and memory retrieval types of recurrency that minimize a specific energy function. In our experiments, we tested these models on artificial and medical imaging data, while analyzing the impact of high levels of noise and few-shot learning settings. Our results do not validate our initial hypothesis that recurrent models should perform better in these settings, suggesting that these recurrent architectures, by themselves, are not sufficient to surpass state-of-the-art feed-forward versions and that additional work needs to be done on the topic., Comment: 24 pages

While convolutional neural networks (CNNs) excel at clean image classification, they struggle to classify images corrupted with different common corruptions, limiting their real-world applicability. Recent work has shown that incorporating a CNN front-end block that simulates some features of the primate primary visual cortex (V1) can improve overall model robustness. Here, we expand on this approach by introducing two novel biologically-inspired CNN model families that incorporate a new front-end block designed to simulate pre-cortical visual processing. RetinaNet, a hybrid architecture containing the novel front-end followed by a standard CNN back-end, shows a relative robustness improvement of 12.3% when compared to the standard model; and EVNet, which further adds a V1 block after the pre-cortical front-end, shows a relative gain of 18.5%. The improvement in robustness was observed for all the different corruption categories, though accompanied by a small decrease in clean image accuracy, and generalized to a different back-end architecture. These findings show that simulating multiple stages of early visual processing in CNN early layers provides cumulative benefits for model robustness.

Modern NLP tasks increasingly rely on dense retrieval methods to access up-to-date and relevant contextual information. We are motivated by the premise that retrieval benefits from segments that can vary in size such that a content's semantic independence is better captured. We propose LumberChunker, a method leveraging an LLM to dynamically segment documents, which iteratively prompts the LLM to identify the point within a group of sequential passages where the content begins to shift. To evaluate our method, we introduce GutenQA, a benchmark with 3000 "needle in a haystack" type of question-answer pairs derived from 100 public domain narrative books available on Project Gutenberg. Our experiments show that LumberChunker not only outperforms the most competitive baseline by 7.37% in retrieval performance (DCG@20) but also that, when integrated into a RAG pipeline, LumberChunker proves to be more effective than other chunking methods and competitive baselines, such as the Gemini 1.5M Pro. Our Code and Data are available at https://github.com/joaodsmarques/LumberChunker

Whole Slide Images (WSI), obtained by high-resolution digital scanning of microscope slides at multiple scales, are the cornerstone of modern Digital Pathology. However, they represent a particular challenge to AI-based/AI-mediated analysis because pathology labeling is typically done at slide-level, instead of tile-level. It is not just that medical diagnostics is recorded at the specimen level, the detection of oncogene mutation is also experimentally obtained, and recorded by initiatives like The Cancer Genome Atlas (TCGA), at the slide level. This configures a dual challenge: a) accurately predicting the overall cancer phenotype and b) finding out what cellular morphologies are associated with it at the tile level. To address these challenges, a weakly supervised Multiple Instance Learning (MIL) approach was explored for two prevalent cancer types, Invasive Breast Carcinoma (TCGA-BRCA) and Lung Squamous Cell Carcinoma (TCGA-LUSC). This approach was explored for tumor detection at low magnification levels and TP53 mutations at various levels. Our results show that a novel additive implementation of MIL matched the performance of reference implementation (AUC 0.96), and was only slightly outperformed by Attention MIL (AUC 0.97). More interestingly from the perspective of the molecular pathologist, these different AI architectures identify distinct sensitivities to morphological features (through the detection of Regions of Interest, RoI) at different amplification levels. Tellingly, TP53 mutation was most sensitive to features at the higher applications where cellular morphology is resolved.

Uncertainty in medical image segmentation tasks, especially inter-rater variability, arising from differences in interpretations and annotations by various experts, presents a significant challenge in achieving consistent and reliable image segmentation. This variability not only reflects the inherent complexity and subjective nature of medical image interpretation but also directly impacts the development and evaluation of automated segmentation algorithms. Accurately modeling and quantifying this variability is essential for enhancing the robustness and clinical applicability of these algorithms. We report the set-up and summarize the benchmark results of the Quantification of Uncertainties in Biomedical Image Quantification Challenge (QUBIQ), which was organized in conjunction with International Conferences on Medical Image Computing and Computer-Assisted Intervention (MICCAI) 2020 and 2021. The challenge focuses on the uncertainty quantification of medical image segmentation which considers the omnipresence of inter-rater variability in imaging datasets. The large collection of images with multi-rater annotations features various modalities such as MRI and CT; various organs such as the brain, prostate, kidney, and pancreas; and different image dimensions 2D-vs-3D. A total of 24 teams submitted different solutions to the problem, combining various baseline models, Bayesian neural networks, and ensemble model techniques. The obtained results indicate the importance of the ensemble models, as well as the need for further research to develop efficient 3D methods for uncertainty quantification methods in 3D segmentation tasks., Comment: initial technical report

How can we detect if copyrighted content was used in the training process of a language model, considering that the training data is typically undisclosed? We are motivated by the premise that a language model is likely to identify verbatim excerpts from its training text. We propose DE-COP, a method to determine whether a piece of copyrighted content was included in training. DE-COP's core approach is to probe an LLM with multiple-choice questions, whose options include both verbatim text and their paraphrases. We construct BookTection, a benchmark with excerpts from 165 books published prior and subsequent to a model's training cutoff, along with their paraphrases. Our experiments show that DE-COP surpasses the prior best method by 9.6% in detection performance (AUC) on models with logits available. Moreover, DE-COP also achieves an average accuracy of 72% for detecting suspect books on fully black-box models where prior methods give approximately 4% accuracy. The code and datasets are available at https://github.com/LeiLiLab/DE-COP.

The ability of large language models (LLMs) to engage in credible dialogues with humans, taking into account the training data and the context of the conversation, has raised discussions about their ability to exhibit intrinsic motivations, agency, or even some degree of consciousness. We argue that the internal architecture of LLMs and their finite and volatile state cannot support any of these properties. By combining insights from complementary learning systems, global neuronal workspace, and attention schema theories, we propose to integrate LLMs and other deep learning systems into an architecture for cognitive language agents able to exhibit properties akin to agency, self-motivation, even some features of meta-cognition.

While some convolutional neural networks (CNNs) have achieved great success in object recognition, they struggle to identify objects in images corrupted with different types of common noise patterns. Recently, it was shown that simulating computations in early visual areas at the front of CNNs leads to improvements in robustness to image corruptions. Here, we further explore this result and show that the neuronal representations that emerge from precisely matching the distribution of RF properties found in primate V1 is key for this improvement in robustness. We built two variants of a model with a front-end modeling the primate primary visual cortex (V1): one sampling RF properties uniformly and the other sampling from empirical biological distributions. The model with the biological sampling has a considerably higher robustness to image corruptions that the uniform variant (relative difference of 8.72%). While similar neuronal sub-populations across the two variants have similar response properties and learn similar downstream weights, the impact on downstream processing is strikingly different. This result sheds light on the origin of the improvements in robustness observed in some biologically-inspired models, pointing to the need of precisely mimicking the neuronal representations found in the primate brain.

Matching addresses is a critical task for companies and post offices involved in the processing and delivery of packages. The ramifications of incorrectly delivering a package to the wrong recipient are numerous, ranging from harm to the company's reputation to economic and environmental costs. This research introduces a deep learning-based model designed to increase the efficiency of address matching for Portuguese addresses. The model comprises two parts: (i) a bi-encoder, which is fine-tuned to create meaningful embeddings of Portuguese postal addresses, utilized to retrieve the top 10 likely matches of the un-normalized target address from a normalized database, and (ii) a cross-encoder, which is fine-tuned to accurately rerank the 10 addresses obtained by the bi-encoder. The model has been tested on a real-case scenario of Portuguese addresses and exhibits a high degree of accuracy, exceeding 95% at the door level. When utilized with GPU computations, the inference speed is about 4.5 times quicker than other traditional approaches such as BM25. An implementation of this system in a real-world scenario would substantially increase the effectiveness of the distribution process. Such an implementation is currently under investigation., Comment: To be published in the 22nd EPIA Conference on Artificial Intelligence, EPIA 2023, Faial Island - Azores, Portugal, 5-8 September 2023, Proceedings

Modern artificial neural networks, including convolutional neural networks and vision transformers, have mastered several computer vision tasks, including object recognition. However, there are many significant differences between the behavior and robustness of these systems and of the human visual system. Deep neural networks remain brittle and susceptible to many changes in the image that do not cause humans to misclassify images. Part of this different behavior may be explained by the type of features humans and deep neural networks use in vision tasks. Humans tend to classify objects according to their shape while deep neural networks seem to rely mostly on texture. Exploring this question is relevant, since it may lead to better performing neural network architectures and to a better understanding of the workings of the vision system of primates. In this work, we advance the state of the art in our understanding of this phenomenon, by extending previous analyses to a much larger set of deep neural network architectures. We found that the performance of models in image classification tasks is highly correlated with their shape bias measured at the output and penultimate layer. Furthermore, our results showed that the number of neurons that represent shape and texture are strongly anti-correlated, thus providing evidence that there is competition between these two types of features. Finally, we observed that while in general there is a correlation between performance and shape bias, there are significant variations between architecture families., Comment: 7 pages, 3 figures

The number of international benchmarking competitions is steadily increasing in various fields of machine learning (ML) research and practice. So far, however, little is known about the common practice as well as bottlenecks faced by the community in tackling the research questions posed. To shed light on the status quo of algorithm development in the specific field of biomedical imaging analysis, we designed an international survey that was issued to all participants of challenges conducted in conjunction with the IEEE ISBI 2021 and MICCAI 2021 conferences (80 competitions in total). The survey covered participants' expertise and working environments, their chosen strategies, as well as algorithm characteristics. A median of 72% challenge participants took part in the survey. According to our results, knowledge exchange was the primary incentive (70%) for participation, while the reception of prize money played only a minor role (16%). While a median of 80 working hours was spent on method development, a large portion of participants stated that they did not have enough time for method development (32%). 25% perceived the infrastructure to be a bottleneck. Overall, 94% of all solutions were deep learning-based. Of these, 84% were based on standard architectures. 43% of the respondents reported that the data samples (e.g., images) were too large to be processed at once. This was most commonly addressed by patch-based training (69%), downsampling (37%), and solving 3D analysis tasks as a series of 2D tasks. K-fold cross-validation on the training set was performed by only 37% of the participants and only 50% of the participants performed ensembling based on multiple identical models (61%) or heterogeneous models (39%). 48% of the respondents applied postprocessing steps.

The Vision Transformer architecture has shown to be competitive in the computer vision (CV) space where it has dethroned convolution-based networks in several benchmarks. Nevertheless, convolutional neural networks (CNN) remain the preferential architecture for the representation module in reinforcement learning. In this work, we study pretraining a Vision Transformer using several state-of-the-art self-supervised methods and assess the quality of the learned representations. To show the importance of the temporal dimension in this context we propose an extension of VICReg to better capture temporal relations between observations by adding a temporal order verification task. Our results show that all methods are effective in learning useful representations and avoiding representational collapse for observations from Atari Learning Environment (ALE) which leads to improvements in data efficiency when we evaluated in reinforcement learning (RL). Moreover, the encoder pretrained with the temporal order verification task shows the best results across all experiments, with richer representations, more focused attention maps and sparser representation vectors throughout the layers of the encoder, which shows the importance of exploring such similarity dimension. With this work, we hope to provide some insights into the representations learned by ViT during a self-supervised pretraining with observations from RL environments and which properties arise in the representations that lead to the best-performing agents. The source code will be available at: https://github.com/mgoulao/TOV-VICReg

The model-based reinforcement learning paradigm, which uses planning algorithms and neural network models, has recently achieved unprecedented results in diverse applications, leading to what is now known as deep reinforcement learning. These agents are quite complex and involve multiple components, factors that can create challenges for research. In this work, we propose a new modular software architecture suited for these types of agents, and a set of building blocks that can be easily reused and assembled to construct new model-based reinforcement learning agents. These building blocks include planning algorithms, policies, and loss functions. We illustrate the use of this architecture by combining several of these building blocks to implement and test agents that are optimized to three different test environments: Cartpole, Minigrid, and Tictactoe. One particular planning algorithm, made available in our implementation and not previously used in reinforcement learning, which we called averaged minimax, achieved good results in the three tested environments. Experiments performed with this architecture have shown that the best combination of planning algorithm, policy, and loss function is heavily problem dependent. This result provides evidence that the proposed architecture, which is modular and reusable, is useful for reinforcement learning researchers who want to study new environments and techniques.

Attention mechanisms have raised significant interest in the research community, since they promise significant improvements in the performance of neural network architectures. However, in any specific problem, we still lack a principled way to choose specific mechanisms and hyper-parameters that lead to guaranteed improvements. More recently, self-attention has been proposed and widely used in transformer-like architectures, leading to significant breakthroughs in some applications. In this work we focus on two forms of attention mechanisms: attention modules and self-attention. Attention modules are used to reweight the features of each layer input tensor. Different modules have different ways to perform this reweighting in fully connected or convolutional layers. The attention models studied are completely modular and in this work they will be used with the popular ResNet architecture. Self-Attention, originally proposed in the area of Natural Language Processing makes it possible to relate all the items in an input sequence. Self-Attention is becoming increasingly popular in Computer Vision, where it is sometimes combined with convolutional layers, although some recent architectures do away entirely with convolutions. In this work, we study and perform an objective comparison of a number of different attention mechanisms in a specific computer vision task, the classification of samples in the widely used Skin Cancer MNIST dataset. The results show that attention modules do sometimes improve the performance of convolutional neural network architectures, but also that this improvement, although noticeable and statistically significant, is not consistent in different settings. The results obtained with self-attention mechanisms, on the other hand, show consistent and significant improvements, leading to the best results even in architectures with a reduced number of parameters.

Medical image segmentation is inherently uncertain. For a given image, there may be multiple plausible segmentation hypotheses, and physicians will often disagree on lesion and organ boundaries. To be suited to real-world application, automatic segmentation systems must be able to capture this uncertainty and variability. Thus far, this has been addressed by building deep learning models that, through dropout, multiple heads, or variational inference, can produce a set - infinite, in some cases - of plausible segmentation hypotheses for any given image. However, in clinical practice, it may not be practical to browse all hypotheses. Furthermore, recent work shows that segmentation variability plateaus after a certain number of independent annotations, suggesting that a large enough group of physicians may be able to represent the whole space of possible segmentations. Inspired by this, we propose a simple method to obtain soft labels from the annotations of multiple physicians and train models that, for each image, produce a single well-calibrated output that can be thresholded at multiple confidence levels, according to each application's precision-recall requirements. We evaluated our method on the MICCAI 2021 QUBIQ challenge, showing that it performs well across multiple medical image segmentation tasks, produces well-calibrated predictions, and, on average, performs better at matching physicians' predictions than other physicians., Comment: 8 pages, 1 figure, 3 tables

Given the inner complexity of the human nervous system, insight into the dynamics of brain activity can be gained from understanding smaller and simpler organisms, such as the nematode C. Elegans. The behavioural and structural biology of these organisms is well-known, making them prime candidates for benchmarking modelling and simulation techniques. In these complex neuronal collections, classical, white-box modelling techniques based on intrinsic structural or behavioural information are either unable to capture the profound nonlinearities of the neuronal response to different stimuli or generate extremely complex models, which are computationally intractable. In this paper we show how the nervous system of C. Elegans can be modelled and simulated with data-driven models using different neural network architectures. Specifically, we target the use of state of the art recurrent neural networks architectures such as LSTMs and GRUs and compare these architectures in terms of their properties and their accuracy as well as the complexity of the resulting models. We show that GRU models with a hidden layer size of 4 units are able to accurately reproduce with high accuracy the system's response to very different stimuli.

Coronary X-ray angiography is a crucial clinical procedure for the diagnosis and treatment of coronary artery disease, which accounts for roughly 16% of global deaths every year. However, the images acquired in these procedures have low resolution and poor contrast, making lesion detection and assessment challenging. Accurate coronary artery segmentation not only helps mitigate these problems, but also allows the extraction of relevant anatomical features for further analysis by quantitative methods. Although automated segmentation of coronary arteries has been proposed before, previous approaches have used non-optimal segmentation criteria, leading to less useful results. Most methods either segment only the major vessel, discarding important information from the remaining ones, or segment the whole coronary tree based mostly on contrast information, producing a noisy output that includes vessels that are not relevant for diagnosis. We adopt a better-suited clinical criterion and segment vessels according to their clinical relevance. Additionally, we simultaneously perform catheter segmentation, which may be useful for diagnosis due to the scale factor provided by the catheter's known diameter, and is a task that has not yet been performed with good results. To derive the optimal approach, we conducted an extensive comparative study of encoder-decoder architectures trained on a combination of focal loss and a variant of generalized dice loss. Based on the EfficientNet and the UNet++ architectures, we propose a line of efficient and high-performance segmentation models using a new decoder architecture, the EfficientUNet++, whose best-performing version achieved average dice scores of 0.8904 and 0.7526 for the artery and catheter classes, respectively, and an average generalized dice score of 0.9234.

Europe was hit hard by the COVID-19 pandemic and Portugal was one of the most affected countries, having suffered three waves in the first twelve months. Approximately between Jan 19th and Feb 5th 2021 Portugal was the country in the world with the largest incidence rate, with 14-days incidence rates per 100,000 inhabitants in excess of 1000. Despite its importance, accurate prediction of the geospatial evolution of COVID-19 remains a challenge, since existing analytical methods fail to capture the complex dynamics that result from both the contagion within a region and the spreading of the infection from infected neighboring regions. We use a previously developed methodology and official municipality level data from the Portuguese Directorate-General for Health (DGS), relative to the first twelve months of the pandemic, to compute an estimate of the incidence rate in each location of mainland Portugal. The resulting sequence of incidence rate maps was then used as a gold standard to test the effectiveness of different approaches in the prediction of the spatial-temporal evolution of the incidence rate. Four different methods were tested: a simple cell level autoregressive moving average (ARMA) model, a cell level vector autoregressive (VAR) model, a municipality-by-municipality compartmental SIRD model followed by direct block sequential simulation and a convolutional sequence-to-sequence neural network model based on the STConvS2S architecture. We conclude that the convolutional sequence-to-sequence neural network is the best performing method, when predicting the medium-term future incidence rate, using the available information., Comment: 10 pages, 8 figures

The significant increase in world population and urbanisation has brought several important challenges, in particular regarding the sustainability, maintenance and planning of urban mobility. At the same time, the exponential increase of computing capability and of available sensor and location data have offered the potential for innovative solutions to these challenges. In this work, we focus on the challenge of traffic forecasting and review the recent development and application of graph neural networks (GNN) to this problem. GNNs are a class of deep learning methods that directly process the input as graph data. This leverages more directly the spatial dependencies of traffic data and makes use of the advantages of deep learning producing state-of-the-art results. We introduce and review the emerging topic of GNNs, including their most common variants, with a focus on its application to traffic forecasting. We address the different ways of modelling traffic forecasting as a (temporal) graph, the different approaches developed so far to combine the graph and temporal learning components, as well as current limitations and research opportunities.

Coronary artery disease leading up to stenosis, the partial or total blocking of coronary arteries, is a severe condition that affects millions of patients each year. Automated identification and classification of stenosis severity from minimally invasive procedures would be of great clinical value, but existing methods do not match the accuracy of experienced cardiologists, due to the complexity of the task. Although a number of computational approaches for quantitative assessment of stenosis have been proposed to date, the performance of these methods is still far from the required levels for clinical applications. In this paper, we propose a two-step deep-learning framework to partially automate the detection of stenosis from X-ray coronary angiography images. In the two steps, we used two distinct convolutional neural network architectures, one to automatically identify and classify the angle of view, and another to determine the bounding boxes of the regions of interest in frames where stenosis is visible. Transfer learning and data augmentation techniques were used to boost the performance of the system in both tasks. We achieved a 0.97 accuracy on the task of classifying the Left/Right Coronary Artery (LCA/RCA) angle view and 0.68/0.73 recall on the determination of the regions of interest, for LCA and RCA, respectively. These results compare favorably with previous results obtained using related approaches, and open the way to a fully automated method for the identification of stenosis severity from X-ray angiographies., Comment: 10 pages, 4 Figures

The combination of deep learning and Monte Carlo Tree Search (MCTS) has shown to be effective in various domains, such as board and video games. AlphaGo represented a significant step forward in our ability to learn complex board games, and it was rapidly followed by significant advances, such as AlphaGo Zero and AlphaZero. Recently, MuZero demonstrated that it is possible to master both Atari games and board games by directly learning a model of the environment, which is then used with MCTS to decide what move to play in each position. During tree search, the algorithm simulates games by exploring several possible moves and then picks the action that corresponds to the most promising trajectory. When training, limited use is made of these simulated games since none of their trajectories are directly used as training examples. Even if we consider that not all trajectories from simulated games are useful, there are thousands of potentially useful trajectories that are discarded. Using information from these trajectories would provide more training data, more quickly, leading to faster convergence and higher sample efficiency. Recent work introduced an off-policy value target for AlphaZero that uses data from simulated games. In this work, we propose a way to obtain off-policy targets using data from simulated games in MuZero. We combine these off-policy targets with the on-policy targets already used in MuZero in several ways, and study the impact of these targets and their combinations in three environments with distinct characteristics. When used in the right combinations, our results show that these targets can speed up the training process and lead to faster convergence and higher rewards than the ones obtained by MuZero.

Gliomas are the most common primary brain malignancies, with different degrees of aggressiveness, variable prognosis and various heterogeneous histologic sub-regions, i.e., peritumoral edematous/invaded tissue, necrotic core, active and non-enhancing core. This intrinsic heterogeneity is also portrayed in their radio-phenotype, as their sub-regions are depicted by varying intensity profiles disseminated across multi-parametric magnetic resonance imaging (mpMRI) scans, reflecting varying biological properties. Their heterogeneous shape, extent, and location are some of the factors that make these tumors difficult to resect, and in some cases inoperable. The amount of resected tumor is a factor also considered in longitudinal scans, when evaluating the apparent tumor for potential diagnosis of progression. Furthermore, there is mounting evidence that accurate segmentation of the various tumor sub-regions can offer the basis for quantitative image analysis towards prediction of patient overall survival. This study assesses the state-of-the-art machine learning (ML) methods used for brain tumor image analysis in mpMRI scans, during the last seven instances of the International Brain Tumor Segmentation (BraTS) challenge, i.e., 2012-2018. Specifically, we focus on i) evaluating segmentations of the various glioma sub-regions in pre-operative mpMRI scans, ii) assessing potential tumor progression by virtue of longitudinal growth of tumor sub-regions, beyond use of the RECIST/RANO criteria, and iii) predicting the overall survival from pre-operative mpMRI scans of patients that underwent gross total resection. Finally, we investigate the challenge of identifying the best ML algorithms for each of these tasks, considering that apart from being diverse on each instance of the challenge, the multi-institutional mpMRI BraTS dataset has also been a continuously evolving/growing dataset., Comment: The International Multimodal Brain Tumor Segmentation (BraTS) Challenge

The Conditional Random Field as a Recurrent Neural Network layer is a recently proposed algorithm meant to be placed on top of an existing Fully-Convolutional Neural Network to improve the quality of semantic segmentation. In this paper, we test whether this algorithm, which was shown to improve semantic segmentation for 2D RGB images, is able to improve segmentation quality for 3D multi-modal medical images. We developed an implementation of the algorithm which works for any number of spatial dimensions, input/output image channels, and reference image channels. As far as we know this is the first publicly available implementation of this sort. We tested the algorithm with two distinct 3D medical imaging datasets, we concluded that the performance differences observed were not statistically significant. Finally, in the discussion section of the paper, we go into the reasons as to why this technique transfers poorly from natural images to medical images., Comment: 11 pages, 3 figures (with 11 subfigures)

The well-known dictionary-based algorithms of the Lempel-Ziv (LZ) 77 family are the basis of several universal lossless compression techniques. These algorithms are asymmetric regarding encoding/decoding time and memory requirements, with the former being much more demanding. In the past years, considerable attention has been devoted to the problem of finding efficient data structures to support these searches, aiming at optimizing the encoders in terms of speed and memory. Hash tables, binary search trees and suffix trees have been widely used for this purpose, as they allow fast search at the expense of memory. Some recent research has focused on suffix arrays (SA), due to their low memory requirements and linear construction algorithms. Previous work has shown how the LZ77 decomposition can be computed using a single SA or an SA with an auxiliary array with the longest common prefix information. The SA-based algorithms use less memory than the tree-based encoders, allocating the strictly necessary amount of memory, regardless of the contents of the text to search/encode. In this paper, we improve on previous work by proposing faster SA-based algorithms for LZ77 encoding and sub-string search, keeping their low memory requirements. For some compression settings, on a large set of benchmark files, our low-memory SA-based encoders are also faster than tree-based encoders. This provides time and memory efficient LZ77 encoding, being a possible replacement for trees on well known encoders like LZMA. Our algorithm is also suited for text classification, because it provides a compact way to describe text in a bag-of-words representation, as well as a fast indexing mechanism that allows to quickly find all the sets of words that start with a given symbol, over a static dictionary., Comment: 10 pages, submitted to DCC2010

Much research has been devoted to optimizing algorithms of the Lempel-Ziv (LZ) 77 family, both in terms of speed and memory requirements. Binary search trees and suffix trees (ST) are data structures that have been often used for this purpose, as they allow fast searches at the expense of memory usage. In recent years, there has been interest on suffix arrays (SA), due to their simplicity and low memory requirements. One key issue is that an SA can solve the sub-string problem almost as efficiently as an ST, using less memory. This paper proposes two new SA-based algorithms for LZ encoding, which require no modifications on the decoder side. Experimental results on standard benchmarks show that our algorithms, though not faster, use 3 to 5 times less memory than the ST counterparts. Another important feature of our SA-based algorithms is that the amount of memory is independent of the text to search, thus the memory that has to be allocated can be defined a priori. These features of low and predictable memory requirements are of the utmost importance in several scenarios, such as embedded systems, where memory is at a premium and speed is not critical. Finally, we point out that the new algorithms are general, in the sense that they are adequate for applications other than LZ compression, such as text retrieval and forward/backward sub-string search., Comment: 10 pages, submited to IEEE - Data Compression Conference 2009