Concept drift detection is a crucial task in data stream evolving environments. Most of state of the art approaches designed to tackle this problem monitor the loss of predictive models. However, this approach falls short in many real-world scenarios, where the true labels are not readily available to compute the loss. In this context, there is increasing attention to approaches that perform concept drift detection in an unsupervised manner, i.e., without access to the true labels. We propose a novel approach to unsupervised concept drift detection based on a student-teacher learning paradigm. Essentially, we create an auxiliary model (student) to mimic the behaviour of the primary model (teacher). At run-time, our approach is to use the teacher for predicting new instances and monitoring the mimicking loss of the student for concept drift detection. In a set of experiments using 19 data streams, we show that the proposed approach can detect concept drift and present a competitive behaviour relative to the state of the art approaches.

Strong empirical evidence that one machine-learning algorithm A outperforms another one B ideally calls for multiple trials optimizing the learning pipeline over sources of variation such as data sampling, data augmentation, parameter initialization, and hyperparameters choices. This is prohibitively expensive, and corners are cut to reach conclusions. We model the whole benchmarking process, revealing that variance due to data sampling, parameter initialization and hyperparameter choice impact markedly the results. We analyze the predominant comparison methods used today in the light of this variance. We show a counter-intuitive result that adding more sources of variation to an imperfect estimator approaches better the ideal estimator at a 51 times reduction in compute cost. Building on these results, we study the error rate of detecting improvements, on five different deep-learning tasks/architectures. This study leads us to propose recommendations for performance comparisons.

This paper presents the first general (supervised) statistical learning framework for point processes in general spaces. Our approach is based on the combination of two new concepts, which we define in the paper: i) bivariate innovations, which are measures of discrepancy/prediction-accuracy between two point processes, and ii) point process cross-validation (CV), which we here define through point process thinning. The general idea is to carry out the fitting by predicting CV-generated validation sets using the corresponding training sets; the prediction error, which we minimise, is measured by means of bivariate innovations. Having established various theoretical properties of our bivariate innovations, we study in detail the case where the CV procedure is obtained through independent thinning and we apply our statistical learning methodology to three typical spatial statistical settings, namely parametric intensity estimation, non-parametric intensity estimation and Papangelou conditional intensity fitting. Aside from deriving theoretical properties related to these cases, in each of them we numerically show that our statistical learning approach outperforms the state of the art in terms of mean (integrated) squared error.

One of the increasingly important technologies dealing with the growing complexity of the digitalization of almost all human activities is Artificial intelligence, more precisely machine learning Despite the fact, that we live in a Big data world where almost everything is digitally stored, there are many real-world situations, where researchers are faced with small data samples. The present study aim is to answer the following research question namely What is the small data problem in machine learning and how it is solved?. Our bibliometric study showed a positive trend in the number of research publications concerning the use of small datasets and substantial growth of the research community dealing with the small dataset problem, indicating that the research field is moving toward higher maturity levels. Despite notable international cooperation, the regional concentration of research literature production in economically more developed countries was observed.

Compared to traditional visual question answering, video-grounded dialogues require additional reasoning over dialogue context to answer questions in a multi-turn setting. Previous approaches to video-grounded dialogues mostly use dialogue context as a simple text input without modelling the inherent information flows at the turn level. In this paper, we propose a novel framework of Reasoning Paths in Dialogue Context (PDC). PDC model discovers information flows among dialogue turns through a semantic graph constructed based on lexical components in each question and answer. PDC model then learns to predict reasoning paths over this semantic graph. Our path prediction model predicts a path from the current turn through past dialogue turns that contain additional visual cues to answer the current question. Our reasoning model sequentially processes both visual and textual information through this reasoning path and the propagated features are used to generate the answer. Our experimental results demonstrate the effectiveness of our method and provide additional insights on how models use semantic dependencies in a dialogue context to retrieve visual cues.

Although depth extraction with passive sensors has seen remarkable improvement with deep learning, these approaches may fail to obtain correct depth if they are exposed to environments not observed during training. Online adaptation, where the neural network trains while deployed, with unsupervised learning provides a convenient solution. However, online adaptation causes a neural network to forget the past. Thus, past training is wasted and the network is not able to provide good results if it observes past scenes. This work deals with practical online-adaptation where the input is online and temporally-correlated, and training is completely unsupervised. Regularization and replay-based methods without task boundaries are proposed to avoid catastrophic forgetting while adapting to online data. Experiments are performed on different datasets with both structure-from-motion and stereo. Results of forgetting as well as adaptation are provided, which are superior to recent methods. The proposed approach is more inline with the artificial general intelligence paradigm as the neural network learns the scene where it is deployed without any supervision (target labels and tasks) and without forgetting about the past. Code is available at github.com/umarKarim/cou_stereo and github.com/umarKarim/cou_sfm.

A few years ago, when I was still working for IBM, I managed an AI project for a bank. During the final phase, my team and I went to the steering committee to present the results. Proud as the project leader, I have shown that the model has achieved 98 percent accuracy in detecting fraudulent transactions. In my manager's eyes, I could see a general panic when I explained that we used an artificial neural network, that it worked with a synapse system and weight adjustments. Although very efficient, there was no way to understand its logic objectively. Even if it was based on real facts, this raw explanation conditioned the project's continuity at that time, unless we could provide a full explanation that the senior executive could understand and trust.

Colby College is carving out space in the liberal arts canon for artificial intelligence. Thanks to a $30 million gift from an alumnus, the small, selective college in Maine is establishing the Davis Institute for Artificial Intelligence, which aims to integrate machine learning, natural language processing and big data into instruction and research across the college. "We want to be sure we're preparing students well for their futures: lives and careers of meaning and purpose," says Margaret McFadden, provost and dean of faculty at Colby. "Well-educated people have to understand AI, what these tools are and how to use them." Artificial intelligence has homes at other U.S. higher ed institutions, including Massachusetts Institute of Technology, the University of Georgia, Stevens Institute of Technology in New Jersey, and Stanford University.

Do you want to upgrade your skills with Best Data Analytics Certification Online to stand out in the industry? Here is a list of Best Data Analytics Courses Online, Training, Tutorials, and Classes to assist you to become a top Data Analyst. Now Big data, Data Science, Machine Learning, Deep Learning, Artificial Intelligence (AI), Analytics, Python, R, r-stats are the most trending and highly demanding subjects in every sector for almost every industry. Learn business analytics to get hands-on knowledge of big data analytics, data visualization, data management, and data mining as an analytics professional. The majority of the business professionals are upgrading their skills with Best Data Analytics Training to standout in their industry.

I don't claim to be a mentor/coach nor do I claim myself to having an extraordinary track record. Although, whatever I am putting down in this blogpost is a result of practical experience that I have over interviewing 100 profiles in the ML domain in last 2–3 years. What we are witnessing today is a flurry of courses in Machine Learning and enormous'interest' in undergraduate students in the pursuing a career in ML. I personally have been approached by numerous undergrads and even some experienced person asking for guidance on how to start with a job in Machine Learning. In this blog, I am consolidating the thoughts and surfacing some myths that a general audience has while starting the journey.