Mauro Dalle Lucca Tosi

Mauro D. L. Tosi, Vinu E. Venugopal, Martin Theobald

Online learning (OL) from data streams is an emerging area of research that encompasses numerous challenges from stream processing, machine learning, and networking. Stream-processing platforms, such as Apache Kafka and Flink, have basic extensions for the training of Artificial Neural Networks (ANNs) in a stream-processing pipeline. However, these extensions were not designed to train ANNs in real-time, and they suffer from performance and scalability issues when doing so. This paper presents TensAIR, the first OL system for training ANNs in real time. TensAIR achieves remarkable performance and scalability by using a decentralized and asynchronous architecture to train ANN models (either freshly initialized or pre-trained) via DASGD (decentralized and asynchronous stochastic gradient descent). We empirically demonstrate that TensAIR achieves a nearly linear scale-out performance in terms of (1) the number of worker nodes deployed in the network, and (2) the throughput at which the data batches arrive at the dataflow operators. We depict the versatility of TensAIR by investigating both sparse (word embedding) and dense (image classification) use cases, for which TensAIR achieved from 6 to 116 times higher sustainable throughput rates than state-of-the-art systems for training ANN in a stream-processing pipeline.

Mauro Dalle Lucca Tosi, Martin Theobald

Online Learning (OL) is a field of research that is increasingly gaining attention both in academia and industry. One of the main challenges of OL is the inherent presence of concept drifts, which are commonly defined as unforeseeable changes in the statistical properties of an incoming data stream over time. The detection of concept drifts typically involves analyzing the error rates produced by an underlying OL algorithm in order to identify if a concept drift occurred or not, such that the OL algorithm can adapt accordingly. Current concept-drift detectors perform very well, i.e., with low false negative rates, but they still tend to exhibit high false positive rates in the concept-drift detection. This may impact the performance of the learner and result in an undue amount of computational resources spent on retraining a model that actually still performs within its expected range. In this paper, we propose OPTWIN, our "OPTimal WINdow" concept drift detector. OPTWIN uses a sliding window of events over an incoming data stream to track the errors of an OL algorithm. The novelty of OPTWIN is to consider both the means and the variances of the error rates produced by a learner in order to split the sliding window into two provably optimal sub-windows, such that the split occurs at the earliest event at which a statistically significant difference according to either the $t$- or the $f$-tests occurred. We assessed OPTWIN over the MOA framework, using ADWIN, DDM, EDDM, STEPD and ECDD as baselines over 7 synthetic and real-world datasets, and in the presence of both sudden and gradual concept drifts. In our experiments, we show that OPTWIN surpasses the F1-score of the baselines in a statistically significant manner while maintaining a lower detection delay and saving up to 21% of time spent on retraining the models.

Henrique X. Goulart, Mauro D. L. Tosi, Daniel Soares Gonçalves et al.

Historically, the Natural Language Processing area has been given too much attention by many researchers. One of the main motivation beyond this interest is related to the word prediction problem, which states that given a set words in a sentence, one can recommend the next word. In literature, this problem is solved by methods based on syntactic or semantic analysis. Solely, each of these analysis cannot achieve practical results for end-user applications. For instance, the Latent Semantic Analysis can handle semantic features of text, but cannot suggest words considering syntactical rules. On the other hand, there are models that treat both methods together and achieve state-of-the-art results, e.g. Deep Learning. These models can demand high computational effort, which can make the model infeasible for certain types of applications. With the advance of the technology and mathematical models, it is possible to develop faster systems with more accuracy. This work proposes a hybrid word suggestion model, based on Naive Bayes and Latent Semantic Analysis, considering neighbouring words around unfilled gaps. Results show that this model could achieve 44.2% of accuracy in the MSR Sentence Completion Challenge.