Country
Synthetic learner: model-free inference on treatments over time
Viviano, Davide, Bradic, Jelena
Understanding of the effect of a particular treatment or a policy pertains to many areas of interest -- ranging from political economics, marketing to health-care and personalized treatment studies. In this paper, we develop a non-parametric, model-free test for detecting the effects of treatment over time that extends widely used Synthetic Control tests. The test is built on counterfactual predictions arising from many learning algorithms. In the Neyman-Rubin potential outcome framework with possible carry-over effects, we show that the proposed test is asymptotically consistent for stationary, beta mixing processes. We do not assume that class of learners captures the correct model necessarily. We also discuss estimates of the average treatment effect, and we provide regret bounds on the predictive performance. To the best of our knowledge, this is the first set of results that allow for example any Random Forest to be useful for provably valid statistical inference in the Synthetic Control setting. In experiments, we show that our Synthetic Learner is substantially more powerful than classical methods based on Synthetic Control or Difference-in-Differences, especially in the presence of non-linear outcome models.
Unveiling phase transitions with machine learning
Canabarro, Askery, Fanchini, Felipe Fernandes, Malvezzi, Andrรฉ Luiz, Pereira, Rodrigo, Chaves, Rafael
The classification of phase transitions is a central and challenging task in condensed matter physics. Typically, it relies on the identification of order parameters and the analysis of singularities in the free energy and its derivatives. Here, we propose an alternative framework to identify quantum phase transitions, employing both unsupervised and supervised machine learning techniques. Using the axial next-nearest neighbor Ising (ANNNI) model as a benchmark, we show how unsupervised learning can detect three phases (ferromagnetic, paramagnetic, and a cluster of the antiphase with the floating phase) as well as two distinct regions within the paramagnetic phase. Employing supervised learning we show that transfer learning becomes possible: a machine trained only with nearest-neighbour interactions can learn to identify a new type of phase occurring when next-nearest-neighbour interactions are introduced. All our results rely on few and low dimensional input data (up to twelve lattice sites), thus providing a computational friendly and general framework for the study of phase transitions in many-body systems.
BCMA-ES: A Bayesian approach to CMA-ES
Benhamou, Eric, Saltiel, David, Verel, Sebastien, Teytaud, Fabien
In a nutshell, the (ยต / ฮป) CMA-ES is an iterative black box optimization algorithm, that, in each of its iterations, samples ฮป candidate This paper introduces a novel theoretically sound approach for solutions from a multivariate normal distribution, evaluates the celebrated CMA-ES algorithm. Assuming the parameters of these solutions (sequentially or in parallel) retains ยต candidates the multi variate normal distribution for the minimum follow a and adjusts the sampling distribution used for the next iteration conjugate prior distribution, we derive their optimal update at to give higher probability to good samples. Each iteration can be each iteration step. Not only provides this Bayesian framework a individually seen as taking an initial guess or prior for the multi justification for the update of the CMA-ES algorithm but it also gives variate parameters, namely the mean and the covariance, and after two new versions of CMA-ES either assuming normal-Wishart or making an experiment by evaluating these sample points with the normal-Inverse Wishart priors, depending whether we parametrize fit function updating the initial parameters accordingly.
On Geometric Structure of Activation Spaces in Neural Networks
Jia, Yuting, Wang, Haiwen, Shao, Shuo, Long, Huan, Zhou, Yunsong, Wang, Xinbing
In this paper, we investigate the geometric structure of activation spaces of fully connected layers in neural networks and then show applications of this study. We propose an efficient approximation algorithm to characterize the convex hull of massive points in high dimensional space. Based on this new algorithm, four common geometric properties shared by the activation spaces are concluded, which gives a rather clear description of the activation spaces. We then propose an alternative classification method grounding on the geometric structure description, which works better than neural networks alone. Surprisingly, this data classification method can be an indicator of overfitting in neural networks. We believe our work reveals several critical intrinsic properties of modern neural networks and further gives a new metric for evaluating them.
UAFS: Uncertainty-Aware Feature Selection for Problems with Missing Data
Becker, Andrew J., Bagrow, James P.
Missing data are a concern in many real world data sets and imputation methods are often needed to estimate the values of missing data, but data sets with excessive missingness and high dimensionality challenge most approaches to imputation. Here we show that appropriate feature selection can be an effective preprocessing step for imputation, allowing for more accurate imputation and subsequent model predictions. The key feature of this preprocessing is that it incorporates uncertainty: by accounting for uncertainty due to missingness when selecting features we can reduce the degree of missingness while also limiting the number of uninformative features being used to make predictive models. We introduce a method to perform uncertainty-aware feature selection (UAFS), provide a theoretical motivation, and test UAFS on both real and synthetic problems, demonstrating that across a variety of data sets and levels of missingness we can improve the accuracy of imputations. Improved imputation due to UAFS also results in improved prediction accuracy when performing supervised learning using these imputed data sets. Our UAFS method is general and can be fruitfully coupled with a variety of imputation methods.
Correlated Parameters to Accurately Measure Uncertainty in Deep Neural Networks
Posch, Konstantin, Pilz, Jรผrgen
In this article a novel approach for training deep neural networks using Bayesian techniques is presented. The Bayesian methodology allows for an easy evaluation of model uncertainty and additionally is robust to overfitting. These are commonly the two main problems classical, i.e. non-Bayesian, architectures have to struggle with. The proposed approach applies variational inference in order to approximate the intractable posterior distribution. In particular, the variational distribution is defined as product of multiple multivariate normal distributions with tridiagonal covariance matrices. Each single normal distribution belongs either to the weights, or to the biases corresponding to one network layer. The layer-wise a posteriori variances are defined based on the corresponding expectation values and further the correlations are assumed to be identical. Therefore, only a few additional parameters need to be optimized compared to non-Bayesian settings. The novel approach is successfully evaluated on basis of the popular benchmark datasets MNIST and CIFAR-10.
Experiments on Open-Set Speaker Identification with Discriminatively Trained Neural Networks
Imoscopi, Stefano, Grancharov, Volodya, Sverrisson, Sigurdur, Karlsson, Erlendur, Pobloth, Harald
This paper presents a study on discriminative artificial neural network classifiers in the context of open-set speaker identification. Both 2-class and multi-class architectures are tested against the conventional Gaussian mixture model based classifier on enrolled speaker sets of different sizes. The performance evaluation shows that the multi-class neural network system has superior performance for large population sizes.
A Survey of Code-switched Speech and Language Processing
Sitaram, Sunayana, Chandu, Khyathi Raghavi, Rallabandi, Sai Krishna, Black, Alan W
Code-switching, the alternation of languages within a conversation or utterance, is a common communicative phenomenon that occurs in multilingual communities across the world. This survey reviews computational approaches for code-switched Speech and Natural Language Processing. We motivate why processing code-switched text and speech is essential for building intelligent agents and systems that interact with users in multilingual communities. As code-switching data and resources are scarce, we list what is available in various code-switched language pairs with the language processing tasks they can be used for. We review code-switching research in various Speech and NLP applications, including language processing tools and end-to-end systems. We conclude with future directions and open problems in the field.
Neural Speed Reading with Structural-Jump-LSTM
Hansen, Christian, Hansen, Casper, Alstrup, Stephen, Simonsen, Jakob Grue, Lioma, Christina
Recurrent neural networks (RNNs) can model natural language by sequentially "reading" input tokens and outputting a distributed representation of each token. Due to the sequential nature of RNNs, inference time is linearly dependent on the input length, and all inputs are read regardless of their importance. Efforts to speed up this inference, known as "neural speed reading", either ignore or skim over part of the input. We present Structural-Jump-LSTM: the first neural speed reading model to both skip and jump text during inference. The model consists of a standard LSTM and two agents: one capable of skipping single words when reading, and one capable of exploiting punctuation structure (sub-sentence separators (,:), sentence end symbols (.!?), or end of text markers) to jump ahead after reading a word. A comprehensive experimental evaluation of our model against all five state-of-the-art neural reading models shows that Structural-Jump-LSTM achieves the best overall floating point operations (FLOP) reduction (hence is faster), while keeping the same accuracy or even improving it compared to a vanilla LSTM that reads the whole text. Recurrent neural networks (RNNs) are a popular model for processing sequential data.
A sparse semismooth Newton based proximal majorization-minimization algorithm for nonconvex square-root-loss regression problems
Tang, Peipei, Wang, Chengjing, Sun, Defeng, Toh, Kim-Chuan
In this paper, we consider high-dimensional nonconvex square-root-loss regression problems and introduce a proximal majorization-minimization (PMM) algorithm for these problems. Our key idea for making the proposed PMM to be efficient is to develop a sparse semismooth Newton method to solve the corresponding subproblems. By using the Kurdyka-{\L}ojasiewicz property exhibited in the underlining problems, we prove that the PMM algorithm converges to a d-stationary point. We also analyze the oracle property of the initial subproblem used in our algorithm. Extensive numerical experiments are presented to demonstrate the high efficiency of the proposed PMM algorithm.