Variational inference is a very efficient and popular heuristic used in various forms in the context of latent variable models. It's closely related to Expectation Maximization (EM), and is applied when exact EM is computationally infeasible. Despite being immensely popular, current theoretical understanding of the effectiveness of variaitonal inference based algorithms is very limited. In this work we provide the first analysis of instances where variational inference algorithms converge to the global optimum, in the setting of topic models.
More specifically, we show that variational inference provably learns the optimal parameters of a topic model under natural assumptions on the topic-word matrix and the topic priors. The properties that the topic word matrix must satisfy in our setting are related to the topic expansion assumption introduced in (Anandkumar et al., 2013), as well as the anchor words assumption in (Arora et al., 2012c). The assumptions on the topic priors are related to the well known Dirichlet prior, introduced to the area of topic modeling by (Blei et al., 2003).
It is well known that initialization plays a crucial role in how well variational based algorithms perform in practice. The initializations that we use are fairly natural. One of them is similar to what is currently used in LDA-c, the most popular implementation of variational inference for topic models. The other one is an overlapping clustering algorithm, inspired by a work by (Arora et al., 2014) on dictionary learning, which is very simple and efficient.
While our primary goal is to provide insights into when variational inference might work in practice, the multiplicative, rather than the additive nature of the variational inference updates forces us to use fairly non-standard proof arguments, which we believe will be of general interest.
Supervised topic models simultaneously model the latent topic structure of large collections of documents and a response variable associated with each document. Existing inference methods are based on variational approximation or Monte Carlo sampling, which often suffers from the local minimum defect. Spectral methods have been applied to learn unsupervised topic models, such as latent Dirichlet allocation (LDA), with provable guarantees. This paper investigates the possibility of applying spectral methods to recover the parameters of supervised LDA (sLDA). We first present a two-stage spectra…
We develop necessary and sufficient conditions and a novel provably consistent and efficient algorithm for discovering topics (latent factors) from observations (documents) that are realized from a probabilistic mixture of shared latent factors that have certain properties. Our focus is on the class of topic models in which each shared latent factor contains a novel word that is unique to that factor, a property that has come to be known as separability. Our algorithm is based on the key insight that the novel words correspond to the extreme points of the convex hull formed by the row-vectors …
University of Lodz（罗兹大学）；Justus Liebig University Giessen（吉森尤斯图斯·李比希大学）
Selecting the number of topics in LDA models is considered to be a difficult task, for which alternative approaches have been proposed. The performance of the recently developed singular Bayesian information criterion (sBIC) is evaluated and compared to the performance of alternative model selection criteria. The sBIC is a generalization of the standard BIC that can be implemented to singular statistical models. The comparison is based on Monte Carlo simulations and carried out for several alternative settings, varying with respect to the number of topics, the number of documents and the size …
We propose a geometric algorithm for topic learning and inference that is built on the convex geometry of topics arising from the Latent Dirichlet Allocation (LDA) model and its nonparametric extensions. To this end we study the optimization of a geometric loss function, which is a surrogate to the LDA's likelihood. Our method involves a fast optimization based weighted clustering procedure augmented with geometric corrections, which overcomes the computational and statistical inefficiencies encountered by other techniques based on Gibbs sampling and variational inference, while achieving the …
Stanford University（斯坦福大学）；Université de Mons（蒙斯大学）
In this paper, we propose a provably correct algorithm for convolutive nonnegative matrix factorization (CNMF) under separability assumptions. CNMF is a convolutive variant of nonnegative matrix factorization (NMF), which functions as an NMF with additional sequential structure. This model is useful in a number of applications, such as audio source separation and neural sequence identification. While a number of heuristic algorithms have been proposed to solve CNMF, to the best of our knowledge no provably correct algorithms have been developed. We present an algorithm that takes advantage of …
Xidian University（西安电子科技大学）；The University of Texas at Austin（得克萨斯大学奥斯汀分校）
To simultaneously capture syntax and global semantics from a text corpus, we propose a new larger-context recurrent neural network (RNN) based language model, which extracts recurrent hierarchical semantic structure via a dynamic deep topic model to guide natural language generation. Moving beyond a conventional RNN-based language model that ignores long-range word dependencies and sentence order, the proposed model captures not only intra-sentence word dependencies, but also temporal transitions between sentences and inter-sentence topic dependencies. For inference, we develop a hybrid of sto…
Indian Institute of Science（印度科学学院）；Microsoft Research（微软研究院）
Topic models, such as Latent Dirichlet Allocation (LDA), posit that documents are drawn from admixtures of distributions over words, known as topics. The inference problem of recovering topics from admixtures, is NP-hard. Assuming separability, a strong assumption, [4] gave the first provable algorithm for inference. For LDA model, [6] gave a provable algorithm using tensor-methods. But [4,6] do not learn topic vectors with bounded $l_1$ error (a natural measure for probability vectors). Our aim is to develop a model which makes intuitive and empirically supported assumptions and to design an …
University of Washington, Seattle（华盛顿大学西雅图分校）；University of Southern California（南加州大学）
The Cheap Gradient Principle (Griewank 2008) --- the computational cost of computing the gradient of a scalar-valued function is nearly the same (often within a factor of $5$) as that of simply computing the function itself --- is of central importance in optimization; it allows us to quickly obtain (high dimensional) gradients of scalar loss functions which are subsequently used in black box gradient-based optimization procedures. The current state of affairs is markedly different with regards to computing subderivatives: widely used ML libraries, including TensorFlow and PyTorch, do not corr…
We describe a layer-by-layer algorithm for training deep convolutional networks, where each step involves gradient updates for a two layer network followed by a simple clustering algorithm. Our algorithm stems from a deep generative model that generates mages level by level, where lower resolution images correspond to latent semantic classes. We analyze the convergence rate of our algorithm assuming that the data is indeed generated according to this model (as well as additional assumptions). While we do not pretend to claim that the assumptions are realistic for natural images, we do believe …
Policy optimization methods remain a powerful workhorse in empirical Reinforcement Learning (RL), with a focus on neural policies that can easily reason over complex and continuous state and/or action spaces. Theoretical understanding of strategic exploration in policy-based methods with non-linear function approximation, however, is largely missing. In this paper, we address this question by designing ENIAC, an actor-critic method that allows non-linear function approximation in the critic. We show that under certain assumptions, e.g., a bounded eluder dimension $d$ for the critic class, the …
Carnegie Mellon University（卡内基梅隆大学）；School of Computer Science（计算机科学学院）；Machine Learning Department（机器学习系）；Forbes Avenue, Pittsburgh, PA 15213, USA（5000 Forbes Avenue, Pittsburgh, PA 15213, USA（地址非机构，此处按要求应忽略，但原文本中该地址与机构关联，需确认））
We consider the problem of matrix column subset selection, which selects a subset of columns from an input matrix such that the input can be well approximated by the span of the selected columns. Column subset selection has been applied to numerous real-world data applications such as population genetics summarization, electronic circuits testing and recommendation systems. In many applications the complete data matrix is unavailable and one needs to select representative columns by inspecting only a small portion of the input matrix. In this paper we propose the first provably correct column …
University of Wisconsin, Madison（威斯康星大学麦迪逊分校）；Rice University（莱斯大学）；University of Washington（华盛顿大学）
We consider interactive learning in the realizable setting and develop a general framework to handle problems ranging from best arm identification to active classification. We begin our investigation with the observation that agnostic algorithms \emph{cannot} be minimax-optimal in the realizable setting. Hence, we design novel computationally efficient algorithms for the realizable setting that match the minimax lower bound up to logarithmic factors and are general-purpose, accommodating a wide variety of function classes including kernel methods, H{ö}lder smooth functions, and convex function…
Artificial Neural Networks (ANNs) have demonstrated remarkable utility in various challenging machine learning applications. While formally verified properties of their behaviors are highly desired, they have proven notoriously difficult to derive and enforce. Existing approaches typically formulate this problem as a post facto analysis process. In this paper, we present a novel learning framework that ensures such formal guarantees are enforced by construction. Our technique enables training provably correct networks with respect to a broad class of safety properties, a capability that goes w…
Department of Mathematics and Operational Research（数学与运筹学系）；Faculté Polytechnique, Université de Mons（蒙斯大学理工学院）
In this paper, we provide novel algorithms with identifiability guarantees for simplex-structured matrix factorization (SSMF), a generalization of nonnegative matrix factorization. Current state-of-the-art algorithms that provide identifiability results for SSMF rely on the sufficiently scattered condition (SSC) which requires the data points to be well spread within the convex hull of the basis vectors. The conditions under which our proposed algorithms recover the unique decomposition is in most cases much weaker than the SSC. We only require to have $d$ points on each facet of the convex hu…
Provably Correct Learning Algorithms in the Presence of Time-Varying Features Using a Variational Perspective
Massachusetts Institute of Technology（麻省理工学院）；Brigham and Women’s Hospital and Harvard Medical School（布里奇沃特医院和哈佛医学院）；Air Force Research Laboratory（空军研究实验室）
In this paper, we consider the problem of training neural network (NN) controllers for nonlinear dynamical systems that are guaranteed to satisfy safety and liveness (e.g., reach-avoid) properties. Our approach is to combine model-based design methodologies for dynamical systems with data-driven approaches to achieve this target. We confine our attention to NNs with Rectifier Linear Unit (ReLU) nonlinearity which are known to represent Continuous Piece-Wise Affine (CPWA) functions. Given a mathematical model of the dynamical system, we compute a finite-state abstract model that captures the cl…
We give a greedy learning algorithm for reconstructing an evolutionary tree based on a certain harmonic average on triplets of terminal taxa. After the pairwise distances between terminal taxa are estimated from sequence data, the algorithm runs in O(n^2) time using O(n) work space, where n is the number of terminal taxa. These time and space complexities are optimal in the sense that the size of an input distance matrix is n^2 and the size of an output tree is n. Moreover, in the Jukes-Cantor model of evolution, the algorithm recovers the correct tree topology with high probability using samp…
University of Texas at Austin（德克萨斯大学奥斯汀分校）；Radboud University Nijmegen（拉德堡德大学奈梅亨分校）
We synthesize shared control protocols subject to probabilistic temporal logic specifications. More specifically, we develop a framework in which a human and an autonomy protocol can issue commands to carry out a certain task. We blend these commands into a joint input to a robot. We model the interaction between the human and the robot as a Markov decision process (MDP) that represents the shared control scenario. Using inverse reinforcement learning, we obtain an abstraction of the human's behavior and decisions. We use randomized strategies to account for randomness in human's decisions, ca…
Non-negative matrix factorization (NMF) is a natural model of admixture and is widely used in science and engineering. A plethora of algorithms have been developed to tackle NMF, but due to the non-convex nature of the problem, there is little guarantee on how well these methods work. Recently a surge of research have focused on a very restricted class of NMFs, called separable NMF, where provably correct algorithms have been developed. In this paper, we propose the notion of subset-separable NMF, which substantially generalizes the property of separability. We show that subset-separability is…
University of Texas at Austin（德克萨斯大学奥斯汀分校）；Tsinghua University（清华大学）；University of Illinois at Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）
Off-policy evaluation provides an essential tool for evaluating the effects of different policies or treatments using only observed data. When applied to high-stakes scenarios such as medical diagnosis or financial decision-making, it is crucial to provide provably correct upper and lower bounds of the expected reward, not just a classical single point estimate, to the end-users, as executing a poor policy can be very costly. In this work, we propose a provably correct method for obtaining interval bounds for off-policy evaluation in a general continuous setting. The idea is to search for the …
We propose a new method of estimation in topic models, that is not a variation on the existing simplex finding algorithms, and that estimates the number of topics K from the observed data. We derive new finite sample minimax lower bounds for the estimation of A, as well as new upper bounds for our proposed estimator. We describe the scenarios where our estimator is minimax adaptive. Our finite sample analysis is valid for any number of documents (n), individual document length (N_i), dictionary size (p) and number of topics (K), and both p and K are allowed to increase with n, a situation not …
A Provably-Correct and Robust Convex Model for Smooth Separable NMF
Department of Mathematics, Hong Kong Baptist University, Hong Kong（香港 Baptist 大学数学系）；Institute of Data Science and Artificial Intelligence, Faculty of Computer Science and Engineering, Innopolis University（Innopolis 大学数据科学与人工智能研究院）；Department of Mathematics and Operational Research, Université de Mons, Rue de Houdain 9, 7000 Mons, Belgium（蒙斯大学数学与运筹系）；European Research Council (ERC consolidator, eLinoR, no 101085607)（欧洲研究理事会（ERC consolidator, eLinoR, no 101085607））
Given a small corpus $\mathcal D_T$ pertaining to a limited set of focused topics, our goal is to train embeddings that accurately capture the sense of words in the topic in spite of the limited size of $\mathcal D_T$. These embeddings may be used in various tasks involving $\mathcal D_T$. A popular strategy in limited data settings is to adapt pre-trained embeddings $\mathcal E$ trained on a large corpus. To correct for sense drift, fine-tuning, regularization, projection, and pivoting have been proposed recently. Among these, regularization informed by a word's corpus frequency performed wel…
Randomized experiments have been critical tools of decision making for decades. However, subjects can show significant heterogeneity in response to treatments in many important applications. Therefore it is not enough to simply know which treatment is optimal for the entire population. What we need is a model that correctly customize treatment assignment base on subject characteristics. The problem of constructing such models from randomized experiments data is known as Uplift Modeling in the literature. Many algorithms have been proposed for uplift modeling and some have generated promising r…
Provably Efficient Off-Policy Adversarial Imitation Learning with Convergence Guarantees
Division of Systems Engineering（系统工程系）；Department of Computer Science（计算机科学系）；Amazon Robotics（亚马逊机器人）；Department of Electrical and Computer Engineering（电气与计算机工程系）
The provably asymptotically fastest algorithm within a factor of 5 for formally described problems will be constructed. The main idea is to enumerate all programs provably equivalent to the original problem by enumerating all proofs. The algorithm could be interpreted as a generalization and improvement of Levin search, which is, within a multiplicative constant, the fastest algorithm for inverting functions. Blum's speed-up theorem is avoided by taking into account only programs for which a correctness proof exists. Furthermore, it is shown that the fastest program that computes a certain fun…
In the optimization of dynamic systems, the variables typically have constraints. Such problems can be modeled as a Constrained Markov Decision Process (CMDP). This paper considers the peak Constrained Markov Decision Process (PCMDP), where the agent chooses the policy to maximize total reward in the finite horizon as well as satisfy constraints at each epoch with probability 1. We propose a model-free algorithm that converts PCMDP problem to an unconstrained problem and a Q-learning based approach is applied. We define the concept of probably approximately correct (PAC) to the proposed PCMDP …
This paper proposes the continuous semantic topic embedding model (CSTEM) which finds latent topic variables in documents using continuous semantic distance function between the topics and the words by means of the variational autoencoder(VAE). The semantic distance could be represented by any symmetric bell-shaped geometric distance function on the Euclidean space, for which the Mahalanobis distance is used in this paper. In order for the semantic distance to perform more properly, we newly introduce an additional model parameter for each word to take out the global factor from this distance …
Self-supervised learning has significantly improved the performance of many NLP tasks. However, how can self-supervised learning discover useful representations, and why is it better than traditional approaches such as probabilistic models are still largely unknown. In this paper, we focus on the context of topic modeling and highlight a key advantage of self-supervised learning - when applied to data generated by topic models, self-supervised learning can be oblivious to the specific model, and hence is less susceptible to model misspecification. In particular, we prove that commonly used sel…
