An official website of the United States government
In point cloud compression, exploiting temporal redundancy for inter predictive coding is challenging because of the irregular geometry. This paper proposes an efficient block-based inter-coding scheme for color attribute compression. The scheme includes integer-precision motion estimation and an adaptive graph based in-loop filtering scheme for improved attribute prediction. The proposed block-based motion estimation scheme consists of an initial motion search that exploits geometric and color attributes, followed by a motion refinement that only minimizes color prediction error. To further improve color prediction, we propose a vertex-domain low-pass graph filtering scheme that can adaptively remove noise from predictors computed from motion estimation with different accuracy. Our experiments demonstrate significant coding gain over state-of-the-art coding methods.
more »
« less
Event Ordering with a Generalized Model for Sieve Prediction Ranking
This paper improves on several aspects of a sieve-based event ordering architecture‚ CAEVO (Chambers et al.‚ 2014)‚ which creates globally consistent temporal relations between events and time expressions. First‚ we examine the usage of word embeddings and semantic role features. With the incorporation of these new features‚ we demonstrate a 5% relative F1 gain over our replicated version of CAEVO. Second‚ we reformulate the architecture’s sieve-based inference algorithm as a prediction reranking method that approximately optimizes a scoring function computed using classifier precisions. Within this prediction reranking framework‚ we propose an alternative scoring function‚ showing an 8.8% relative gain over the original CAEVO. We further include an in-depth analysis of one of the main datasets that is used to evaluate temporal classifiers‚ and we show that in spite of the density of this corpus‚ there is still a danger of overfitting. While this paper focuses on temporal ordering‚ its results are applicable to other areas that use sievebased architectures.
more »
« less
- Award ID(s):
- 1528409
- PAR ID:
- 10067554
- Date Published:
- Journal Name:
- Proceedings of the 8th International Joint Conference on Natural Language Processing
- Page Range / eLocation ID:
- 843-853
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
The mmWave WiGig frequency band can support high throughput and low latency emerging applications. In this context, accurate prediction of channel gain enables seamless connectivity with user mobility via proactive handover and beamforming. Machine learning techniques have been widely adopted in literature for mmWave channel prediction. However, the existing techniques assume that the indoor mmWave channel follows a stationary stochastic process. This paper demonstrates that indoor WiGig mmWave channels are non-stationary where the channel’s cumulative distribution function (CDF) changes with the user’s spatio-temporal mobility. Specifically, we show significant differences in the empirical CDF of the channel gain based on the user’s mobility stage, namely, room entering, wandering, and exiting. Thus, the dynamic WiGig mmWave indoor channel suffers from concept drift that impedes the generalization ability of deep learning-based channel prediction models. Our results demonstrate that a state-of-the-art deep learning channel prediction model based on a hybrid convolutional neural network (CNN) long-short-term memory (LSTM) recurrent neural network suffers from a deterioration in the prediction accuracy by 11–68% depending on the user’s mobility stage and the model’s training. To mitigate the negative effect of concept drift and improve the generalization ability of the channel prediction model, we develop a robust deep learning model based on an ensemble strategy. Our results show that the weight average ensemble-based model maintains a stable prediction that keeps the performance deterioration below 4%.more » « less
-
null (Ed.)Peer prediction mechanisms incentivize agents to truthfully report their signals even in the absence of verification by comparing agents’ reports with those of their peers. In the detail-free multi-task setting, agents are asked to respond to multiple independent and identically distributed tasks, and the mechanism does not know the prior distribution of agents’ signals. The goal is to provide an epsilon-strongly truthful mechanism where truth-telling rewards agents “strictly” more than any other strategy profile (with epsilon additive error) even for heterogeneous agents, and to do so while requiring as few tasks as possible. We design a family of mechanisms with a scoring function that maps a pair of reports to a score. The mechanism is strongly truthful if the scoring function is “prior ideal”. Moreover, the mechanism is epsilon-strongly truthful as long as the scoring function used is sufficiently close to the ideal scoring function. This reduces the above mechanism design problem to a learning problem – specifically learning an ideal scoring function. Because learning the prior distribution is sufficient (but not necessary) to learn the scoring function, we can apply standard learning theory techniques that leverage side information about the prior (e.g., that it is close to some parametric model). Furthermore, we derive a variational representation of an ideal scoring function and reduce the learning problem into an empirical risk minimization. We leverage this reduction to obtain very general results for peer prediction in the multi-task setting. Specifically, Sample Complexity. We show how to derive good bounds on the number of tasks required for different types of priors–in some cases exponentially improving previous results. In particular, we can upper bound the required number of tasks for parametric models with bounded learning complexity. Furthermore, our reduction applies to myriad continuous signal space settings. To the best of our knowledge, this is the first peer-prediction mechanism on continuous signals designed for the multi-task setting. Connection to Machine Learning. We show how to turn a soft-predictor of an agent’s signals (given the other agents’ signals) into a mechanism. This allows the practical use of machine learning algorithms that give good results even when many agents provide noisy information. Stronger Properties. In the finite setting, we obtain -strongly truthful mechanisms for any stochastically relevant prior. Prior works either only apply to more restrictive settings, or achieve a weaker notion of truthfulness (informed truthfulness).more » « less
-
Abstract The geometric sieve for densities is a very convenient tool proposed by Poonen and Stoll (and independently by Ekedahl) to compute the density of a given subset of the integers. In this paper we provide an effective criterion to find all higher moments of the density (e.g. the mean, the variance) of a subset of a finite dimensional free module over the ring of algebraic integers of a number field. More precisely, we provide a geometric sieve that allows the computation of all higher moments corresponding to the density, over a general number fieldK. This work advances the understanding of geometric sieve for density computations in two ways: on one hand, it extends a result of Bright, Browning and Loughran, where they provide the geometric sieve for densities over number fields; on the other hand, it extends the recent result on a geometric sieve for expected values over the integers to both the ring of algebraic integers and to moments higher than the expected value. To show how effective and applicable our method is, we compute the density, mean and variance of Eisenstein polynomials and shifted Eisenstein polynomials over number fields. This extends (and fully covers) results in the literature that were obtained with ad-hoc methods.more » « less
-
The predictive Intensive Care Unit (ICU) scoring system plays an important role in ICU management for its capability of predicting important outcomes, especially mortality. There are many scoring systems that have been developed and used in the ICU. These scoring systems are primarily based on the structured clinical data contained in the electronic health record (EHR), which may suffer the loss of the important clinical information contained in the narratives and images. In this work, we build a deep learning based survival prediction model with multimodality data to predict ICU-mortality. Four sets of features are investigated: (1) physiological measurements of Simplified Acute Physiology Score (SAPS) II, (2) common thorax diseases predefined by radiologists, (3) BERT-based text representations, and (4) chest X-ray image features. We use the Medical Information Mart for Intensive Care IV (MIMIC-IV) dataset to evaluate the proposed model. Our model achieves the average C-index of 0.7847 (95% confidence interval, 0.7625–0.8068), which substantially exceeds that of the baseline with SAPS-II features (0.7477 (0.7238–0.7716)). Ablation studies further demonstrate the contributions of pre-defined labels (2.12%), text features (2.68%), and image features (2.96%). Our model achieves a higher average C-index than the traditional machine learning methods under the same feature fusion setting, which suggests that the deep learning methods can outperform the traditional machine learning methods in ICU-mortality prediction. These results highlight the potential of deep learning models with multimodal information to enhance ICU-mortality prediction. We make our work publicly available at https://github.com/bionlplab/mimic-icu-mortality.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- The DOI is not currently available.
