Machine Learning
Deep Reinforcement Learning with Feedback-based Exploration (1903.06151v1)
Jan Scholten, Daan Wout, Carlos Celemin, Jens Kober
2019-03-14
Deep Reinforcement Learning has enabled the control of increasingly complex and high-dimensional problems. However, the need of vast amounts of data before reasonable performance is attained prevents its widespread application. We employ binary corrective feedback as a general and intuitive manner to incorporate human intuition and domain knowledge in model-free machine learning. The uncertainty in the policy and the corrective feedback is combined directly in the action space as probabilistic conditional exploration. As a result, the greatest part of the otherwise ignorant learning process can be avoided. We demonstrate the proposed method, Predictive Probabilistic Merging of Policies (PPMP), in combination with DDPG. In experiments on continuous control problems of the OpenAI Gym, we achieve drastic improvements in sample efficiency, final performance, and robustness to erroneous feedback, both for human and synthetic feedback. Additionally, we show solutions beyond the demonstrated knowledge.
Decision Variance in Online Learning (1807.09089v2)
Sattar Vakili, Alexis Boukouvalas, Qing Zhao
2018-07-24
Online learning has traditionally focused on the expected rewards. In this paper, a risk-averse online learning problem under the performance measure of the mean-variance of the rewards is studied. Both the bandit and full information settings are considered. The performance of several existing policies is analyzed, and new fundamental limitations on risk-averse learning is established. In particular, it is shown that although a logarithmic distribution-dependent regret in time
is achievable (similar to the risk-neutral problem), the worst-case (i.e. minimax) regret is lower bounded by (in contrast to the lower bound in the risk-neutral problem). This sharp difference from the risk-neutral counterpart is caused by the the variance in the player's decisions, which, while absent in the regret under the expected reward criterion, contributes to excess mean-variance due to the non-linearity of this risk measure. The role of the decision variance in regret performance reflects a risk-averse player's desire for robust decisions and outcomes.
Online Learning with Diverse User Preferences (1901.07924v3)
Chao Gan, Jing Yang, Ruida Zhou, Cong Shen
2019-01-23
In this paper, we investigate the impact of diverse user preference on learning under the stochastic multi-armed bandit (MAB) framework. We aim to show that when the user preferences are sufficiently diverse and each arm can be optimal for certain users, the O(log T) regret incurred by exploring the sub-optimal arms under the standard stochastic MAB setting can be reduced to a constant. Our intuition is that to achieve sub-linear regret, the number of times an optimal arm being pulled should scale linearly in time; when all arms are optimal for certain users and pulled frequently, the estimated arm statistics can quickly converge to their true values, thus reducing the need of exploration dramatically. We cast the problem into a stochastic linear bandits model, where both the users preferences and the state of arms are modeled as {independent and identical distributed (i.i.d)} d-dimensional random vectors. After receiving the user preference vector at the beginning of each time slot, the learner pulls an arm and receives a reward as the linear product of the preference vector and the arm state vector. We also assume that the state of the pulled arm is revealed to the learner once its pulled. We propose a Weighted Upper Confidence Bound (W-UCB) algorithm and show that it can achieve a constant regret when the user preferences are sufficiently diverse. The performance of W-UCB under general setups is also completely characterized and validated with synthetic data.
Deep Switch Networks for Generating Discrete Data and Language (1903.06135v1)
Payam Delgosha, Naveen Goela
2019-03-14
Multilayer switch networks are proposed as artificial generators of high-dimensional discrete data (e.g., binary vectors, categorical data, natural language, network log files, and discrete-valued time series). Unlike deconvolution networks which generate continuous-valued data and which consist of upsampling filters and reverse pooling layers, multilayer switch networks are composed of adaptive switches which model conditional distributions of discrete random variables. An interpretable, statistical framework is introduced for training these nonlinear networks based on a maximum-likelihood objective function. To learn network parameters, stochastic gradient descent is applied to the objective. This direct optimization is stable until convergence, and does not involve back-propagation over separate encoder and decoder networks, or adversarial training of dueling networks. While training remains tractable for moderately sized networks, Markov-chain Monte Carlo (MCMC) approximations of gradients are derived for deep networks which contain latent variables. The statistical framework is evaluated on synthetic data, high-dimensional binary data of handwritten digits, and web-crawled natural language data. Aspects of the model's framework such as interpretability, computational complexity, and generalization ability are discussed.
Tucker Tensor Layer in Fully Connected Neural Networks (1903.06133v1)
Giuseppe G. Calvi, Ahmad Moniri, Mahmoud Mahfouz, Zeyang Yu, Qibin Zhao, Danilo P. Mandic
2019-03-14
We introduce the Tucker Tensor Layer (TTL), an alternative to the dense weight-matrices of the fully connected layers of feed-forward neural networks (NNs), to answer the long standing quest to compress NNs and improve their interpretability. This is achieved by treating these weight-matrices as the unfolding of a higher order weight-tensor. This enables us to introduce a framework for exploiting the multi-way nature of the weight-tensor in order to efficiently reduce the number of parameters, by virtue of the compression properties of tensor decompositions. The Tucker Decomposition (TKD) is employed to decompose the weight-tensor into a core tensor and factor matrices. We re-derive back-propagation within this framework, by extending the notion of matrix derivatives to tensors. In this way, the physical interpretability of the TKD is exploited to gain insights into training, through the process of computing gradients with respect to each factor matrix. The proposed framework is validated on synthetic data and on the Fashion-MNIST dataset, emphasizing the relative importance of various data features in training, hence mitigating the "black-box" issue inherent to NNs. Experiments on both MNIST and Fashion-MNIST illustrate the compression properties of the TTL, achieving a 66.63 fold compression whilst maintaining comparable performance to the uncompressed NN.
Adversarial Bandits with Knapsacks (1811.11881v3)
Nicole Immorlica, Karthik Abinav Sankararaman, Robert Schapire, Aleksandrs Slivkins
2018-11-28
We consider Bandits with Knapsacks (henceforth, BwK), a general model for multi-armed bandits under supply/budget constraints. In particular, a bandit algorithm needs to solve a well-known knapsack problem: find an optimal packing of items into a limited-size knapsack. The BwK problem is a common generalization of numerous motivating examples, which range from dynamic pricing to repeated auctions to dynamic ad allocation to network routing and scheduling. While the prior work on BwK focused on the stochastic version, we pioneer the other extreme in which the outcomes can be chosen adversarially. This is a considerably harder problem, compared to both the stochastic version and the "classic" adversarial bandits, in that regret minimization is no longer feasible. Instead, the objective is to minimize the competitive ratio: the ratio of the benchmark reward to the algorithm's reward. We design an algorithm with competitive ratio O(log T) relative to the best fixed distribution over actions, where T is the time horizon; we also prove a matching lower bound. The key conceptual contribution is a new perspective on the stochastic version of the problem. We suggest a new algorithm for the stochastic version, which builds on the framework of regret minimization in repeated games and admits a substantially simpler analysis compared to prior work. We then analyze this algorithm for the adversarial version and use it as a subroutine to solve the latter.
The Diverse Cohort Selection Problem (1709.03441v5)
Candice Schumann, Samsara N. Counts, Jeffrey S. Foster, John P. Dickerson
2017-09-11
How should a firm allocate its limited interviewing resources to select the optimal cohort of new employees from a large set of job applicants? How should that firm allocate cheap but noisy resume screenings and expensive but in-depth in-person interviews? We view this problem through the lens of combinatorial pure exploration (CPE) in the multi-armed bandit setting, where a central learning agent performs costly exploration of a set of arms before selecting a final subset with some combinatorial structure. We generalize a recent CPE algorithm to the setting where arm pulls can have different costs and return different levels of information. We then prove theoretical upper bounds for a general class of arm-pulling strategies in this new setting. We apply our general algorithm to a real-world problem with combinatorial structure: incorporating diversity into university admissions. We take real data from admissions at one of the largest US-based computer science graduate programs and show that a simulation of our algorithm produces a cohort with hiring overall utility while spending comparable budget to the current admissions process at that university.
Separating the EoR Signal with a Convolutional Denoising Autoencoder: A Deep-learning-based Method (1902.09278v2)
Weitian Li, Haiguang Xu, Zhixian Ma, Ruimin Zhu, Dan Hu, Zhenghao Zhu, Junhua Gu, Chenxi Shan, Jie Zhu, Xiang-Ping Wu
2019-02-25
When applying the foreground removal methods to uncover the faint cosmological signal from the epoch of reionization (EoR), the foreground spectra are assumed to be smooth. However, this assumption can be seriously violated in practice since the unresolved or mis-subtracted foreground sources, which are further complicated by the frequency-dependent beam effects of interferometers, will generate significant fluctuations along the frequency dimension. To address this issue, we propose a novel deep-learning-based method that uses a 9-layer convolutional denoising autoencoder (CDAE) to separate the EoR signal. After being trained on the SKA images simulated with realistic beam effects, the CDAE achieves excellent performance as the mean correlation coefficient (
) between the reconstructed and input EoR signals reaches . In comparison, the two representative traditional methods, namely the polynomial fitting method and the continuous wavelet transform method, both have difficulties in modelling and removing the foreground emission complicated with the beam effects, yielding only and , respectively. We conclude that, by hierarchically learning sophisticated features through multiple convolutional layers, the CDAE is a powerful tool that can be used to overcome the complicated beam effects and accurately separate the EoR signal. Our results also exhibit the great potential of deep-learning-based methods in future EoR experiments.
High-dimensional Index Volatility Models via Stein's Identity (1811.10790v2)
Sen Na, Mladen Kolar
2018-11-27
We study estimation of the parametric components of single and multiple index volatility models. Using the first- and second-order Stein's identity, we develop methods that are applicable for estimation of the variance index in a high-dimensional setting requiring finite moment condition, which allows for heavy-tailed data. Our approach complements the existing literature in a low-dimensional setting, while relaxing the conditions on estimation, and provides a novel approach in a high-dimensional setting. We prove that the statistical rate of convergence of our variance index estimators consists of a parametric rate and a nonparametric rate, where the latter appears from the estimation of the mean link function. However, under standard assumptions, the parametric rate dominates the rate of convergence and our results match the minimax optimal rate for the mean index estimation. Simulation results illustrate finite sample properties of our methodology and back our theoretical conclusions.
Federated Learning in Distributed Medical Databases: Meta-Analysis of Large-Scale Subcortical Brain Data (1810.08553v3)
Santiago Silva, Boris Gutman, Eduardo Romero, Paul M Thompson, Andre Altmann, Marco Lorenzi
2018-10-19
At this moment, databanks worldwide contain brain images of previously unimaginable numbers. Combined with developments in data science, these massive data provide the potential to better understand the genetic underpinnings of brain diseases. However, different datasets, which are stored at different institutions, cannot always be shared directly due to privacy and legal concerns, thus limiting the full exploitation of big data in the study of brain disorders. Here we propose a federated learning framework for securely accessing and meta-analyzing any biomedical data without sharing individual information. We illustrate our framework by investigating brain structural relationships across diseases and clinical cohorts. The framework is first tested on synthetic data and then applied to multi-centric, multi-database studies including ADNI, PPMI, MIRIAD and UK Biobank, showing the potential of the approach for further applications in distributed analysis of multi-centric cohorts