Showing papers on "Algorithmic learning theory published in 2017"

Deep Bayesian active learning with image data

Abstract: Even though active learning forms an important pillar of machine learning, deep learning tools are not prevalent within it. Deep learning poses several difficulties when used in an active learning setting. First, active learning (AL) methods generally rely on being able to learn and update models from small amounts of data. Recent advances in deep learning, on the other hand, are notorious for their dependence on large amounts of data. Second, many AL acquisition functions rely on model uncertainty, yet deep learning methods rarely represent such model uncertainty. In this paper we combine recent advances in Bayesian deep learning into the active learning framework in a practical way. We develop an active learning framework for high dimensional data, a task which has been extremely challenging so far, with very sparse existing literature. Taking advantage of specialised models such as Bayesian convolutional neural networks, we demonstrate our active learning techniques with image data, obtaining a significant improvement on existing active learning approaches. We demonstrate this on both the MNIST dataset, as well as for skin cancer diagnosis from lesion images (ISIC2016 task).

A Proposal on Machine Learning via Dynamical Systems

Abstract: We discuss the idea of using continuous dynamical systems to model general high-dimensional nonlinear functions used in machine learning. We also discuss the connection with deep learning.

The need to approximate the use-case in clinical machine learning.

Abstract: The availability of smartphone and wearable sensor technology is leading to a rapid accumulation of human subject data, and machine learning is emerging as a technique to map those data into clinical predictions. As machine learning algorithms are increasingly used to support clinical decision making, it is vital to reliably quantify their prediction accuracy. Cross-validation (CV) is the standard approach where the accuracy of such algorithms is evaluated on part of the data the algorithm has not seen during training. However, for this procedure to be meaningful, the relationship between the training and the validation set should mimic the relationship between the training set and the dataset expected for the clinical use. Here we compared two popular CV methods: record-wise and subject-wise. While the subject-wise method mirrors the clinically relevant use-case scenario of diagnosis in newly recruited subjects, the record-wise strategy has no such interpretation. Using both a publicly available dataset and a simulation, we found that record-wise CV often massively overestimates the prediction accuracy of the algorithms. We also conducted a systematic review of the relevant literature, and found that this overly optimistic method was used by almost half of the retrieved studies that used accelerometers, wearable sensors, or smartphones to predict clinical outcomes. As we move towards an era of machine learning-based diagnosis and treatment, using proper methods to evaluate their accuracy is crucial, as inaccurate results can mislead both clinicians and data scientists.

Guest Column: A Survey of Quantum Learning Theory

Abstract: This paper surveys quantum learning theory: the theoretical aspects of machine learning using quantum computers. We describe the main results known for three models of learning: exact learning from membership queries, and Probably Approximately Correct (PAC) and agnostic learning from classical or quantum examples.

Parallel learning: a perspective and a framework

Abstract: The development of machine learning in complex system is hindered by two problems nowadays. The first problem is the inefficiency of exploration in state and action space, which leads to the data-hungry of some state-of-art data-driven algorithm. The second problem is the lack of a general theory which can be used to analyze and implement a complex learning system. In this paper, we proposed a general methods that can address both two issues. We combine the concepts of descriptive learning, predictive learning, and prescriptive learning into a uniform framework, so as to build a parallel system allowing learning system improved by self-boosting. Formulating a new perspective of data, knowledge and action, we provide a new methodology called parallel learning to design machine learning system for real-world problems.

Machine Teaching: A New Paradigm for Building Machine Learning Systems.

Abstract: The current processes for building machine learning systems require practitioners with deep knowledge of machine learning. This significantly limits the number of machine learning systems that can be created and has led to a mismatch between the demand for machine learning systems and the ability for organizations to build them. We believe that in order to meet this growing demand for machine learning systems we must significantly increase the number of individuals that can teach machines. We postulate that we can achieve this goal by making the process of teaching machines easy, fast and above all, universally accessible. While machine learning focuses on creating new algorithms and improving the accuracy of "learners", the machine teaching discipline focuses on the efficacy of the "teachers". Machine teaching as a discipline is a paradigm shift that follows and extends principles of software engineering and programming languages. We put a strong emphasis on the teacher and the teacher's interaction with data, as well as crucial components such as techniques and design principles of interaction and visualization. In this paper, we present our position regarding the discipline of machine teaching and articulate fundamental machine teaching principles. We also describe how, by decoupling knowledge about machine learning algorithms from the process of teaching, we can accelerate innovation and empower millions of new uses for machine learning models.

A Survey on Machine Learning: Concept,Algorithms and Applications

Abstract: Over the past few decades, Machine Learning (ML) has evolved from the endeavour of few computer enthusiasts exploiting the possibility of computers learning to play games, and a part of Mathematics (Statistics) that seldom considered computational approaches, to an independent research discipline that has not only provided the necessary base for statistical-computational principles of learning procedures, but also has developedvarious algorithms that are regularly used for text interpretation, pattern recognition, and a many other commercial purposes and has led to a separate research interest in data mining to identify hidden regularities or irregularities in social data that growing by second. This paper focuses on explaining the concept and evolution of Machine Learning, some of the popular Machine Learning algorithms and try to compare three most popular algorithms based on some basic notions. Sentiment140 dataset was used and performance of each algorithm in terms of training time, prediction time and accuracy of prediction have been documented and compared.

On learning and branching: a survey

Abstract: This paper surveys learning techniques to deal with the two most crucial decisions in the branch-and-bound algorithm for Mixed-Integer Linear Programming, namely variable and node selections. Because of the lack of deep mathematical understanding on those decisions, the classical and vast literature in the field is inherently based on computational studies and heuristic, often problem-specific, strategies. We will both interpret some of those early contributions in the light of modern (machine) learning techniques, and give the details of the recent algorithms that instead explicitly incorporate machine learning paradigms.

Online Transfer Learning with Multiple Homogeneous or Heterogeneous Sources

Abstract: Transfer learning techniques have been broadly applied in applications where labeled data in a target domain are difficult to obtain while a lot of labeled data are available in related source domains. In practice, there can be multiple source domains that are related to the target domain, and how to combine them is still an open problem. In this paper, we seek to leverage labeled data from multiple source domains to enhance classification performance in a target domain where the target data are received in an online fashion. This problem is known as the online transfer learning problem. To achieve this, we propose novel online transfer learning paradigms in which the source and target domains are leveraged adaptively. We consider two different problem settings: homogeneous transfer learning and heterogeneous transfer learning. The proposed methods work in an online manner, where the weights of the source domains are adjusted dynamically. We provide the mistake bounds of the proposed methods and perform comprehensive experiments on real-world data sets to demonstrate the effectiveness of the proposed algorithms.

Learning style Identifier: Improving the precision of learning style identification through computational intelligence algorithms

Abstract: Identifying students’ learning styles has several benefits such as making students aware of their strengths and weaknesses when it comes to learning and the possibility to personalize their learning environment to their learning styles. While there exist learning style questionnaires for identifying a student's learning style, such questionnaires have several disadvantages and therefore, research has been conducted on automatically identifying learning styles from students’ behavior in a learning environment. Current approaches to automatically identify learning styles have an average precision between 66% and 77%, which shows the need for improvements in order to use such automatic approaches reliably in learning environments. In this paper, four computational intelligence algorithms (artificial neural network, genetic algorithm, ant colony system and particle swarm optimization) have been investigated with respect to their potential to improve the precision of automatic learning style identification. Each algorithm was evaluated with data from 75 students. The artificial neural network shows the most promising results with an average precision of 80.7%, followed by particle swarm optimization with an average precision of 79.1%. Improving the precision of automatic learning style identification allows more students to benefit from more accurate information about their learning styles as well as more accurate personalization towards accommodating their learning styles in a learning environment. Furthermore, teachers can have a better understanding of their students and be able to provide more appropriate interventions.

Quantum machine learning: a classical perspective

Abstract: Recently, increased computational power and data availability, as well as algorithmic advances, have led machine learning techniques to impressive results in regression, classification, data-generation and reinforcement learning tasks. Despite these successes, the proximity to the physical limits of chip fabrication alongside the increasing size of datasets are motivating a growing number of researchers to explore the possibility of harnessing the power of quantum computation to speed-up classical machine learning algorithms. Here we review the literature in quantum machine learning and discuss perspectives for a mixed readership of classical machine learning and quantum computation experts. Particular emphasis will be placed on clarifying the limitations of quantum algorithms, how they compare with their best classical counterparts and why quantum resources are expected to provide advantages for learning problems. Learning in the presence of noise and certain computationally hard problems in machine learning are identified as promising directions for the field. Practical questions, like how to upload classical data into quantum form, will also be addressed.

What's statistical about learning? Insights from modelling statistical learning as a set of memory processes

Abstract: Statistical learning has been studied in a variety of different tasks, including word segmentation, object identification, category learning, artificial grammar learning and serial reaction time tasks (e.g. Saffran et al. 1996 Science 274 , 1926–1928; Orban et al. 2008 Proceedings of the National Academy of Sciences 105 , 2745–2750; Thiessen & Yee 2010 Child Development 81 , 1287–1303; Saffran 2002 Journal of Memory and Language 47 , 172–196; Misyak & Christiansen 2012 Language Learning 62 , 302–331). The difference among these tasks raises questions about whether they all depend on the same kinds of underlying processes and computations, or whether they are tapping into different underlying mechanisms. Prior theoretical approaches to statistical learning have often tried to explain or model learning in a single task. However, in many cases these approaches appear inadequate to explain performance in multiple tasks. For example, explaining word segmentation via the computation of sequential statistics (such as transitional probability) provides little insight into the nature of sensitivity to regularities among simultaneously presented features. In this article, we will present a formal computational approach that we believe is a good candidate to provide a unifying framework to explore and explain learning in a wide variety of statistical learning tasks. This framework suggests that statistical learning arises from a set of processes that are inherent in memory systems, including activation, interference, integration of information and forgetting (e.g. Perruchet & Vinter 1998 Journal of Memory and Language 39 , 246–263; Thiessen et al. 2013 Psychological Bulletin 139 , 792–814). From this perspective, statistical learning does not involve explicit computation of statistics, but rather the extraction of elements of the input into memory traces, and subsequent integration across those memory traces that emphasize consistent information (Thiessen and Pavlik 2013 Cognitive Science 37 , 310–343). This article is part of the themed issue ‘New frontiers for statistical learning in the cognitive sciences'.

Comparative study of machine learning techniques in sentimental analysis

Abstract: Sentimental Analysis is reference to the task of Natural Language Processing to determine whether a text contains subjective information and what information it expresses i.e., whether the attitude behind the text is positive, negative or neutral. This paper focuses on the several machine learning techniques which are used in analyzing the sentiments and in opinion mining. Sentimental analysis with the blend of machine learning could be useful in predicting the product reviews and consumer attitude towards to newly launched product. This paper presents a detail survey of various machine learning techniques and then compared with their accuracy, advantages and limitations of each technique. On comparing we get 85% of accuracy by using supervised machine learning technique which is higher than that of unsupervised learning techniques.

Complexity vs. performance: empirical analysis of machine learning as a service

Abstract: Machine learning classifiers are basic research tools used in numerous types of network analysis and modeling. To reduce the need for domain expertise and costs of running local ML classifiers, network researchers can instead rely on centralized Machine Learning as a Service (MLaaS) platforms. In this paper, we evaluate the effectiveness of MLaaS systems ranging from fully-automated, turnkey systems to fully-customizable systems, and find that with more user control comes greater risk. Good decisions produce even higher performance, and poor decisions result in harsher performance penalties. We also find that server side optimizations help fully-automated systems outperform default settings on competitors, but still lag far behind well-tuned MLaaS systems which compare favorably to standalone ML libraries. Finally, we find classifier choice is the dominating factor in determining model performance, and that users can approximate the performance of an optimal classifier choice by experimenting with a small subset of random classifiers. While network researchers should approach MLaaS systems with caution, they can achieve results comparable to standalone classifiers if they have sufficient insight into key decisions like classifiers and feature selection.

An Incremental Type-2 Meta-Cognitive Extreme Learning Machine

Abstract: Existing extreme learning algorithm have not taken into account four issues: 1) complexity; 2) uncertainty; 3) concept drift; and 4) high dimensionality. A novel incremental type-2 meta-cognitive extreme learning machine (ELM) called evolving type-2 ELM (eT2ELM) is proposed to cope with the four issues in this paper. The eT2ELM presents three main pillars of human meta-cognition: 1) what-to-learn; 2) how-to-learn; and 3) when-to-learn. The what-to-learn component selects important training samples for model updates by virtue of the online certainty-based active learning method, which renders eT2ELM as a semi-supervised classifier. The how-to-learn element develops a synergy between extreme learning theory and the evolving concept, whereby the hidden nodes can be generated and pruned automatically from data streams with no tuning of hidden nodes. The when-to-learn constituent makes use of the standard sample reserved strategy. A generalized interval type-2 fuzzy neural network is also put forward as a cognitive component, in which a hidden node is built upon the interval type-2 multivariate Gaussian function while exploiting a subset of Chebyshev series in the output node. The efficacy of the proposed eT2ELM is numerically validated in 12 data streams containing various concept drifts. The numerical results are confirmed by thorough statistical tests, where the eT2ELM demonstrates the most encouraging numerical results in delivering reliable prediction, while sustaining low complexity.

Using Human Brain Activity to Guide Machine Learning

Abstract: Machine learning is a field of computer science that builds algorithms that learn. In many cases, machine learning algorithms are used to recreate a human ability like adding a caption to a photo, driving a car, or playing a game. While the human brain has long served as a source of inspiration for machine learning, little effort has been made to directly use data collected from working brains as a guide for machine learning algorithms. Here we demonstrate a new paradigm of "neurally-weighted" machine learning, which takes fMRI measurements of human brain activity from subjects viewing images, and infuses these data into the training process of an object recognition learning algorithm to make it more consistent with the human brain. After training, these neurally-weighted classifiers are able to classify images without requiring any additional neural data. We show that our neural-weighting approach can lead to large performance gains when used with traditional machine vision features, as well as to significant improvements with already high-performing convolutional neural network features. The effectiveness of this approach points to a path forward for a new class of hybrid machine learning algorithms which take both inspiration and direct constraints from neuronal data.

Limits of End-to-End Learning

Abstract: End-to-end learning refers to training a possibly complex learning system by applying gradient-based learning to the system as a whole. End-to-end learning system is specifically designed so that all modules are differentiable. In effect, not only a central learning machine, but also all "peripheral" modules like representation learning and memory formation are covered by a holistic learning process. The power of end-to-end learning has been demonstrated on many tasks, like playing a whole array of Atari video games with a single architecture. While pushing for solutions to more challenging tasks, network architectures keep growing more and more complex. In this paper we ask the question whether and to what extent end-to-end learning is a future-proof technique in the sense of scaling to complex and diverse data processing architectures. We point out potential inefficiencies, and we argue in particular that end-to-end learning does not make optimal use of the modular design of present neural networks. Our surprisingly simple experiments demonstrate these inefficiencies, up to the complete breakdown of learning.

Unsupervised feature selection via low-rank approximation and structure learning

Abstract: Feature selection is an important research topic in machine learning and computer vision in that it can reduce the dimensionality of input data and improve the performance of learning algorithms. Low-rank approximation techniques can well exploit the low-rank property of input data, which coincides with the internal consistency of dimensionality reduction. In this paper, we propose an efficient unsupervised feature selection algorithm, which incorporates low-rank approximation as well as structure learning. First, using the self-representation of data matrix, we formalize the feature selection problem as a matrix factorization with low-rank constraints. This matrix factorization formulation also embeds structure learning regularization as well as a sparse regularized term. Second, we present an effective technique to approximate low-rank constraints and propose a convergent algorithm in a batch mode. This technique can serve as an algorithmic framework for general low-rank recovery problems as well. Finally, the proposed algorithm is validated in twelve publicly available datasets from machine learning repository. Extensive experimental results demonstrate that the proposed method is capable to achieve competitive performance compared to existing state-of-the-art feature selection methods in terms of clustering performance.

Distributed Multi-Task Relationship Learning

Abstract: Multi-task learning aims to learn multiple tasks jointly by exploiting their relatedness to improve the generalization performance for each task. Traditionally, to perform multi-task learning, one needs to centralize data from all the tasks to a single machine. However, in many real-world applications, data of different tasks may be geo-distributed over different local machines. Due to heavy communication caused by transmitting the data and the issue of data privacy and security, it is impossible to send data of different task to a master machine to perform multi-task learning. Therefore, in this paper, we propose a distributed multi-task learning framework that simultaneously learns predictive models for each task as well as task relationships between tasks alternatingly in the parameter server paradigm. In our framework, we first offer a general dual form for a family of regularized multi-task relationship learning methods. Subsequently, we propose a communication-efficient primal-dual distributed optimization algorithm to solve the dual problem by carefully designing local subproblems to make the dual problem decomposable. Moreover, we provide a theoretical convergence analysis for the proposed algorithm, which is specific for distributed multi-task relationship learning. We conduct extensive experiments on both synthetic and real-world datasets to evaluate our proposed framework in terms of effectiveness and convergence.

Large Scale Graph Learning from Smooth Signals

Abstract: Graphs are a prevalent tool in data science, as they model the inherent structure of the data. They have been used successfully in unsupervised and semi-supervised learning. Typically they are constructed either by connecting nearest samples, or by learning them from data, solving an optimization problem. While graph learning does achieve a better quality, it also comes with a higher computational cost. In particular, the current state-of-the-art model cost is $\mathcal{O}(n^2)$ for $n$ samples. In this paper, we show how to scale it, obtaining an approximation with leading cost of $\mathcal{O}(n\log(n))$, with quality that approaches the exact graph learning model. Our algorithm uses known approximate nearest neighbor techniques to reduce the number of variables, and automatically selects the correct parameters of the model, requiring a single intuitive input: the desired edge density.

Model-based learning: a synthesis of theory and research

Abstract: This article provides a review of theoretical approaches to model-based learning and related research. In accordance with the definition of model-based learning as an acquisition and utilization of mental models by learners, the first section centers on mental model theory. In accordance with epistemology of modeling the issues of semantics, ontology, and learning with models as well as structural aspects and functions of mental models (such as simplification, idealization, analogy, simulation) will be discussed. Starting with early theoretical approaches in the realm of cognitive science, the emphasis will be on recent theoretical developments in the field of reinforcement learning that distinguish between model-based and model-free learning. These new theoretical approaches confirm, to a large extent, the early theories of mental models but they also contribute new insights in age-related aspects of model-based learning. The second main section of the article provides a review of basic and applied research on model-based learning. A short overview of basic research on mental models aims at sketching an “overall picture” followed by a more detailed description and discussion of findings in the field of instructional research, especially with regard to K-12 STEM education and the support of model-based learning through teaching and technology.

Image recognition method based on deep learning

Abstract: Deep learning algorithms are a subset of the machine learning algorithms, which aim at discovering multiple levels of distributed representations. Recently, numerous deep learning algorithms have been proposed to solve traditional artificial intelligence problems. This work aims to review the state-of-the-art in deep learning algorithms in computer vision by highlighting the contributions and challenges from recent research papers. It first gives an overview of various deep learning approaches and their recent developments, and then briefly describes their applications in diverse vision tasks. Finally, the paper summarizes the future trends and challenges in designing and training deep neural networks.

Understanding Grounded Language Learning Agents

Abstract: Neural network-based systems can now learn to locate the referents of words and phrases in images, answer questions about visual scenes, and even execute symbolic instructions as first-person actors in partially-observable worlds. To achieve this so-called grounded language learning, models must overcome certain well-studied learning challenges that are also fundamental to infants learning their first words. While it is notable that models with no meaningful prior knowledge overcome these learning obstacles, AI researchers and practitioners currently lack a clear understanding of exactly how they do so. Here we address this question as a way of achieving a clearer general understanding of grounded language learning, both to inform future research and to improve confidence in model predictions. For maximum control and generality, we focus on a simple neural network-based language learning agent trained via policy-gradient methods to interpret synthetic linguistic instructions in a simulated 3D world. We apply experimental paradigms from developmental psychology to this agent, exploring the conditions under which established human biases and learning effects emerge. We further propose a novel way to visualise and analyse semantic representation in grounded language learning agents that yields a plausible computational account of the observed effects.

Efficient learning of constraints and generic null space policies

Abstract: A large class of motions can be decomposed into a movement task and null-space policy subject to a set of constraints. When learning such motions from demonstrations, we aim to achieve generalisation across different unseen constraints and to increase the robustness to noise while keeping the computational cost low. There exists a variety of methods for learning the movement policy and the constraints. The effectiveness of these techniques has been demonstrated in low-dimensional scenarios and simple motions. In this paper, we present a fast and accurate approach to learning constraints from observations. This novel formulation of the problem allows the constraint learning method to be coupled with the policy learning method to improve policy learning accuracy, which enables us to learn more complex motions. We demonstrate our approach by learning a complex surface wiping policy in a 7-DOF robotic arm.

Learning behavior and achievement analysis of a digital game-based learning approach integrating mastery learning theory and different feedback models

Abstract: It is widely accepted that the digital game-based learning approach has the advantage of stimulating students’ learning motivation, but simply using digital games in the classroom does not guarantee satisfactory learning achievement, especially in the case of the absence of a teacher. Integrating appropriate learning strategies into a game can better enhance the learning performance. Therefore, in this study, a mastery theory-based digital game with different feedback models was developed to compare the differences in the learning behavior of students using the two feedback models. Lag sequential analysis was then applied to identify the sequential behaviors that are statistically proven to have impact. The results of the experiments and behavior analysis show that, with proper design of the game, students in both feedback methods can achieve the same learning performance as that in the conventional learning method with a teacher involved. Moreover, students in the Regular Feedback Group reviewed ...

Learning how to Active Learn: A Deep Reinforcement Learning Approach

Abstract: Active learning aims to select a small subset of data for annotation such that a classifier learned on the data is highly accurate. This is usually done using heuristic selection methods, however the effectiveness of such methods is limited and moreover, the performance of heuristics varies between datasets. To address these shortcomings, we introduce a novel formulation by reframing the active learning as a reinforcement learning problem and explicitly learning a data selection policy, where the policy takes the role of the active learning heuristic. Importantly, our method allows the selection policy learned using simulation on one language to be transferred to other languages. We demonstrate our method using cross-lingual named entity recognition, observing uniform improvements over traditional active learning.