Search results

Extreme multi-label classification (XMC) aims at assigning to an instance the most relevant subset of labels from a colossal label set. There have been some success in formulating the multi-label problem as sequence-to-sequence (Seq2Seq) learning, where the positive class labels of each input instance are used as the corresponding output sequence. Seq2Seq methods, nonetheless, have not yet been scalable

We derive algorithms for producing optimal aggregation schedules for automatically aggregating gradients across di!erent compute units, both CPUs and GPUs, with arbitrary topologies. We show that this can be accomplished by solving a linear program on the spanning tree polytope. We give analytic bounds for the value of the optimal solution for arbitrary graphs. We also propose simple schedules that meet

Faced with distribution shift between training and test set, we wish to detect and quantify the shift, and to correct our classifiers without test set labels. Motivated by medical diagnosis, where diseases (targets), cause symptoms (observations), we focus on label shift, where the label marginal p(y) changes but the conditional p(x|y) does not. We propose Black Box Shift Estimation (BBSE) to estimate the

Deploying deep learning (DL) models on edge devices is getting popular nowadays. The huge diversity of edge devices, with both computation and memory constraints, however, make efficient deployment challenging. In this paper, we propose a two-stage pipeline that optimizes DL models on target devices. The first stage optimizes the inference workloads, and the second stage searches optimal kernel implementations

We consider the problem of labeling points on a fast-moving data stream when only a small number of labeled examples are available. In our setting, incoming points must be processed efficiently and the stream is too large to store in its entirety. We present a semi-supervised learning algorithm for this task. The algorithm maintains a small synopsis of the stream which can be quickly updated as new points

We consider the task of predicting subjective fashion traits from images. Specifically, we are interested in understanding which outfit actually better suites the user. Since these traits are highly subjective, they tend to be noisier. One solution is to annotate each example several times, but this makes it hard to collect large amounts of data.

Recent advances in automatic speech recognition lead toward enabling a voice conversation between a human user and an intelligent virtual assistant. This provides a potential foundation for developing artificial personal shoppers for e-commerce websites, such as Alibaba, Amazon, and eBay. Personal shoppers are valuable to the on-line shops as they enhance user engagement and trust by promptly dealing with

We consider the online sparse linear regression problem, which is the problem of sequentially making predictions observing only a limited number of features in each round, to minimize regret with respect to the best sparse linear regressor, where prediction accuracy is measured by square loss. We give an inefficient algorithm that obtains regret bounded by O˜( √ T) after T prediction rounds. We complement

This article presents a whisper speech detector in the far-field domain. The proposed system consists of a long short-term memory (LSTM) neural network trained on log-filterbank energy (LFBE) acoustic features. This model is trained and evaluated on recordings of human interactions with voice-controlled, far-field devices in whisper and normal phonation modes. We compare multiple inference approaches for

Tree-structured LSTMs have shown advantages in learning semantic representations by exploiting syntactic information. Most existing methods model tree structures by bottomup combinations of constituent nodes using the same shared compositional function and often making use of input word information only. The inability to capture the richness of compositionality makes these models lack expressive power.

Managing risk is important to any E-commerce merchant. Various machine learning (ML) models combined with a rule set as the decision layer is a common practice to manage the risks. Unlike the ML models that can be automatically refreshed periodically based on new risk patterns, rules are generally static and rely on manual updates. To tackle that, this paper presents a data-driven and automated rule optimization

Recent speech synthesis systems based on sampling from autoregressive neural networks models can generate speech almost undistinguishable from human recordings. However, these models require large amounts of data. This paper shows that the lack of data from one speaker can be compensated with data from other speakers. The naturalness of Tacotron2-like models trained on a blend of 5k utterances from 7 speakers

Multi-fidelity methods are prominently used when cheaply-obtained, but possibly biased and noisy, observations must be effectively combined with limited or expensive true data in order to construct reliable models. This arises in both fundamental machine learning procedures such as Bayesian optimization, as well as more practical science and engineering applications. In this paper we develop a novel multi-fidelity

Deep neural networks are often brittle to superficial perturbations of their inputs; models that perform well offline on held-out data can still break under small amounts of naturally-occurring or adversarial shifts. We consider invariant representation learning (IRL), first proposed in the domain of speech recognition, as a simple, effective, and general extension to data augmentation. Rather than only

Goal-oriented dialogue systems typically rely on components specifically developed for a single task or domain. This limits such systems in two different ways: If there is an update in the task domain, the dialogue system usually needs to be updated or completely re-trained. It is also harder to extend such dialogue systems to different and multiple domains. The dialogue state tracker in conventional dialogue

We propose an alternative training framework for Bayesian neural networks (BNNs), which is motivated by viewing the Bayesian model for supervised learning as an autoencoder for data transmission. Then, a natural objective can be invoked from the rate-distortion theory. Specifically, we end up minimizing the mutual information between the weights and the dataset with a constraint that the negative log-likelihood

In domain classification for spoken dialog systems, correct detection of out-of-domain (OOD) utterances is crucial because it reduces confusion and unnecessary interaction costs between users and the systems. Previous work usually utilizes OOD detectors that are trained separately from in-domain (IND) classifiers, and confidence thresholding for OOD detection given target evaluation scores. In this paper

Building conversational systems that enable natural language interactions with machines has been attractive to mankind since the early days of computing, as exemplified by earlier text-based systems such as ELIZA [Weizenbaum, 1966]. Previous work on conversational systems generally falls into two categories, task-oriented and socialbots. Task-oriented systems aim to help users accomplish a specific task

In September 2016, Amazon introduced a new problem in conversational Al, the Alexa Prize, which challenged student teams to build a socialbot that could converse with Alexa on a wide range of topics. The socialbot differs from task-oriented dialog systems that address explicit user goals associated with a constrained domain, and also differs from chatbot systems that only handle social chitchat. The socialbot

This paper presents the second version of the dialogue system named Alquist competing in Amazon Alexa Prize 2018. We introduce a system leveraging ontology based topic structure called topic nodes. Each of the nodes consists of several sub-dialogues, and each sub-dialogue has its own LSTM-based model for dialogue management. The sub-dialogues can be triggered according to the topic hierarchy or a user intent