Machine Learning
Categorical Feature Compression via Submodular Optimization (1904.13389v1)
MohammadHossein Bateni, Lin Chen, Hossein Esfandiari, Thomas Fu, Vahab S. Mirrokni, Afshin Rostamizadeh
2019-04-30
In the era of big data, learning from categorical features with very large vocabularies (e.g., 28 million for the Criteo click prediction dataset) has become a practical challenge for machine learning researchers and practitioners. We design a highly-scalable vocabulary compression algorithm that seeks to maximize the mutual information between the compressed categorical feature and the target binary labels and we furthermore show that its solution is guaranteed to be within a
factor of the global optimal solution. To achieve this, we introduce a novel re-parametrization of the mutual information objective, which we prove is submodular, and design a data structure to query the submodular function in amortized time (where is the input vocabulary size). Our complete algorithm is shown to operate in time. Additionally, we design a distributed implementation in which the query data structure is decomposed across machines such that each machine only requires space, while still preserving the approximation guarantee and using only logarithmic rounds of computation. We also provide analysis of simple alternative heuristic compression methods to demonstrate they cannot achieve any approximation guarantee. Using the large-scale Criteo learning task, we demonstrate better performance in retaining mutual information and also verify competitive learning performance compared to other baseline methods.
Risk-Averse Explore-Then-Commit Algorithms for Finite-Time Bandits (1904.13387v1)
Ali Yekkehkhany, Ebrahim Arian, Mohammad Hajiesmaili, Rakesh Nagi
2019-04-30
In this paper, we study multi-armed bandit problems in explore-then-commit setting. In our proposed explore-then-commit setting, the goal is to identify the best arm after a pure experimentation (exploration) phase and exploit it once or for a given finite number of times. We identify that although the arm with the highest expected reward is the most desirable objective for infinite exploitations, it is not necessarily the one that is most probable to have the highest reward in a single or finite-time exploitations. Alternatively, we advocate the idea of risk-aversion where the objective is to compete against the arm with the best risk-return trade-off. Then, we propose two algorithms whose objectives are to select the arm that is most probable to reward the most. Using a new notion of finite-time exploitation regret, we find an upper bound for the minimum number of experiments before commitment, to guarantee an upper bound for the regret. As compared to existing risk-averse bandit algorithms, our algorithms do not rely on hyper-parameters, resulting in a more robust behavior in practice, which is verified by the numerical evaluation.
Active Manifolds: A non-linear analogue to Active Subspaces (1904.13386v1)
Robert A. Bridges, Anthony D. Gruber, Cristopher R. Felder, Miki Verma, Chelsey Hoff
2019-04-30
We present an approach to analyze
functions that addresses limitations present in the Active Subspaces (AS) method of Constantine et al.(2015; 2014). Under appropriate hypotheses, our Active Manifolds (AM) method identifies a 1-D curve in the domain (the active manifold) on which nearly all values of the unknown function are attained, and which can be exploited for approximation or analysis, especially when is large (high-dimensional in-put space). We provide theorems justifying our AM technique and an algorithm permitting functional approximation and sensitivity analysis. Using accessible, low-dimensional functions as initial examples, we show AM reduces approximation error by an order of magnitude compared to AS, at the expense of more computation. Following this, we revisit the sensitivity analysis by Glaws et al. (2017), who apply AS to analyze a magnetohydrodynamic power generator model, and compare the performance of AM on the same data. Our analysis provides detailed information not captured by AS, exhibiting the influence of each parameter individually along an active manifold. Overall, AM represents a novel technique for analyzing functional models with benefits including: reducing -dimensional analysis to a 1-D analogue, permit-ting more accurate regression than AS (at more computational expense), enabling more informative sensitivity analysis, and granting accessible visualizations(2-D plots) of parameter sensitivity along the AM.
Distillation Strategies for Proximal Policy Optimization (1901.08128v2)
Sam Green, Craig M. Vineyard, Çetin Kaya Koç
2019-01-23
Vision-based deep reinforcement learning (RL) typically obtains performance benefit by using high capacity and relatively large convolutional neural networks (CNN). However, a large network leads to higher inference costs (power, latency, silicon area, MAC count). Many inference optimizations have been developed for CNNs. Some optimization techniques offer theoretical efficiency, such as sparsity, but designing actual hardware to support them is difficult. On the other hand, distillation is a simple general-purpose optimization technique which is broadly applicable for transferring knowledge from a trained, high capacity teacher network to an untrained, low capacity student network. DQN distillation extended the original distillation idea to transfer information stored in a high performance, high capacity teacher Q-function trained via the Deep Q-Learning (DQN) algorithm. Our work adapts the DQN distillation work to the actor-critic Proximal Policy Optimization algorithm. PPO is simple to implement and has much higher performance than the seminal DQN algorithm. We show that a distilled PPO student can attain far higher performance compared to a DQN teacher. We also show that a low capacity distilled student is generally able to outperform a low capacity agent that directly trains in the environment. Finally, we show that distillation, followed by "fine-tuning" in the environment, enables the distilled PPO student to achieve parity with teacher performance. In general, the lessons learned in this work should transfer to other modern actor-critic RL algorithms.
FRI - Feature Relevance Intervals for Interpretable and Interactive Data Exploration (1903.00719v2)
Lukas Pfannschmidt, Christina Göpfert, Ursula Neumann, Dominik Heider, Barbara Hammer
2019-03-02
Most existing feature selection methods are insufficient for analytic purposes as soon as high dimensional data or redundant sensor signals are dealt with since features can be selected due to spurious effects or correlations rather than causal effects. To support the finding of causal features in biomedical experiments, we hereby present FRI, an open source Python library that can be used to identify all-relevant variables in linear classification and (ordinal) regression problems. Using the recently proposed feature relevance method, FRI is able to provide the base for further general experimentation or in specific can facilitate the search for alternative biomarkers. It can be used in an interactive context, by providing model manipulation and visualization methods, or in a batch process as a filter method.
Very Deep Self-Attention Networks for End-to-End Speech Recognition (1904.13377v1)
Ngoc-Quan Pham, Thai-Son Nguyen, Jan Niehues, Markus Muller, Alex Waibel
2019-04-30
Recently, end-to-end sequence-to-sequence models for speech recognition have gained significant interest in the research community. While previous architecture choices revolve around time-delay neural networks (TDNN) and long short-term memory (LSTM) recurrent neural networks, we propose to use self-attention via the Transformer architecture as an alternative. Our analysis shows that deep Transformer networks with high learning capacity are able to exceed performance from previous end-to-end approaches and even match the conventional hybrid systems. Moreover, we trained very deep models with up to 48 Transformer layers for both encoder and decoders combined with stochastic residual connections, which greatly improve generalizability and training efficiency. The resulting models outperform all previous end-to-end ASR approaches on the Switchboard benchmark. An ensemble of these models achieve 9.9% and 17.7% WER on Switchboard and CallHome test sets respectively. This finding brings our end-to-end models to competitive levels with previous hybrid systems. Further, with model ensembling the Transformers can outperform certain hybrid systems, which are more complicated in terms of both structure and training procedure.
Gradient Coding Based on Block Designs for Mitigating Adversarial Stragglers (1904.13373v1)
Swanand Kadhe, O. Ozan Koyluoglu, Kannan Ramchandran
2019-04-30
Distributed implementations of gradient-based methods, wherein a server distributes gradient computations across worker machines, suffer from slow running machines, called 'stragglers'. Gradient coding is a coding-theoretic framework to mitigate stragglers by enabling the server to recover the gradient sum in the presence of stragglers. 'Approximate gradient codes' are variants of gradient codes that reduce computation and storage overhead per worker by allowing the server to approximately reconstruct the gradient sum. In this work, our goal is to construct approximate gradient codes that are resilient to stragglers selected by a computationally unbounded adversary. Our motivation for constructing codes to mitigate adversarial stragglers stems from the challenge of tackling stragglers in massive-scale elastic and serverless systems, wherein it is difficult to statistically model stragglers. Towards this end, we propose a class of approximate gradient codes based on balanced incomplete block designs (BIBDs). We show that the approximation error for these codes depends only on the number of stragglers, and thus, adversarial straggler selection has no advantage over random selection. In addition, the proposed codes admit computationally efficient decoding at the server. Next, to characterize fundamental limits of adversarial straggling, we consider the notion of 'adversarial threshold' -- the smallest number of workers that an adversary must straggle to inflict certain approximation error. We compute a lower bound on the adversarial threshold, and show that codes based on symmetric BIBDs maximize this lower bound among a wide class of codes, making them excellent candidates for mitigating adversarial stragglers.
Factor Analysis in Fault Diagnostics Using Random Forest (1904.13366v1)
Nagdev Amruthnath, Tarun Gupta
2019-04-30
Factor analysis or sometimes referred to as variable analysis has been extensively used in classification problems for identifying specific factors that are significant to particular classes. This type of analysis has been widely used in application such as customer segmentation, medical research, network traffic, image, and video classification. Today, factor analysis is prominently being used in fault diagnosis of machines to identify the significant factors and to study the root cause of a specific machine fault. The advantage of performing factor analysis in machine maintenance is to perform prescriptive analysis (helps answer what actions to take?) and preemptive analysis (helps answer how to eliminate the failure mode?). In this paper, a real case of an industrial rotating machine was considered where vibration and ambient temperature data was collected for monitoring the health of the machine. Gaussian mixture model-based clustering was used to cluster the data into significant groups, and spectrum analysis was used to diagnose each cluster to a specific state of the machine. The significant features that attribute to a particular mode of the machine were identified by using the random forest classification model. The significant features for specific modes of the machine were used to conclude that the clusters generated are distinct and have a unique set of significant features.
The Level Weighted Structural Similarity Loss: A Step Away from the MSE (1904.13362v1)
Yingjing Lu
2019-04-30
The Mean Square Error (MSE) has shown its strength when applied in deep generative models such as Auto-Encoders to model reconstruction loss. However, in image domain especially, the limitation of MSE is obvious: it assumes pixel independence and ignores spatial relationships of samples. This contradicts most architectures of Auto-Encoders which use convolutional layers to extract spatial dependent features. We base on the structural similarity metric (SSIM) and propose a novel level weighted structural similarity (LWSSIM) loss for convolutional Auto-Encoders. Experiments on common datasets on various Auto-Encoder variants show that our loss is able to outperform the MSE loss and the Vanilla SSIM loss. We also provide reasons why our model is able to succeed in cases where the standard SSIM loss fails.
Structured Prediction using cGANs with Fusion Discriminator (1904.13358v1)
Faisal Mahmood, Wenhao Xu, Nicholas J. Durr, Jeremiah W. Johnson, Alan Yuille
2019-04-30
We propose the fusion discriminator, a single unified framework for incorporating conditional information into a generative adversarial network (GAN) for a variety of distinct structured prediction tasks, including image synthesis, semantic segmentation, and depth estimation. Much like commonly used convolutional neural network -- conditional Markov random field (CNN-CRF) models, the proposed method is able to enforce higher-order consistency in the model, but without being limited to a very specific class of potentials. The method is conceptually simple and flexible, and our experimental results demonstrate improvement on several diverse structured prediction tasks.