PyTorchKR Papers

최신 논문

총 715편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2205.11169공식 코드 · Papers with Code

PEVL: Position-enhanced Pre-training and Prompt Tuning for Vision-language Models

Vision-language pre-training (VLP) has shown impressive performance on a wide range of cross-modal tasks, where VLP models without reliance on object detectors are becoming the mainstream due to their superior computation efficiency and competitive performance. However, the removal of object detectors also deprives the capability of VLP models in explicit object modeling, which is essential to various position-sensitive vision-language (VL) tasks, such as referring expression comprehension and visual commonsense reasoning. To address the challenge, we introduce PEVL that enhances the pre-training and prompt tuning of VLP models with explicit object position modeling. Specifically, PEVL reformulates discretized object positions and language in a unified language modeling framework, which facilitates explicit VL alignment during pre-training, and also enables flexible prompt tuning for various downstream tasks. We show that PEVL enables state-of-the-art performance of detector-free VLP models on position-sensitive tasks such as referring expression comprehension and phrase grounding, and also improves the performance on position-insensitive tasks with grounded inputs. We make the data and code for this paper publicly available at https://github.com/thunlp/PEVL.

arXiv:2205.09921공식 코드 · Papers with Code

KERPLE: Kernelized Relative Positional Embedding for Length Extrapolation

Relative positional embeddings (RPE) have received considerable attention since RPEs effectively model the relative distance among tokens and enable length extrapolation. We propose KERPLE, a framework that generalizes relative position embedding for extrapolation by kernelizing positional differences. We achieve this goal using conditionally positive definite (CPD) kernels, a class of functions known for generalizing distance metrics. To maintain the inner product interpretation of self-attention, we show that a CPD kernel can be transformed into a PD kernel by adding a constant offset. This offset is implicitly absorbed in the Softmax normalization during self-attention. The diversity of CPD kernels allows us to derive various RPEs that enable length extrapolation in a principled way. Experiments demonstrate that the logarithmic variant achieves excellent extrapolation performance on three large language modeling datasets. Our implementation and pretrained checkpoints are released at~\url{https://github.com/chijames/KERPLE.git}.

arXiv:2205.09054공식 코드 · Papers with Code

Position Aided Beam Prediction in the Real World: How Useful GPS Locations Actually Are?

Millimeter-wave (mmWave) communication systems rely on narrow beams for achieving sufficient receive signal power. Adjusting these beams is typically associated with large training overhead, which becomes particularly critical for highly-mobile applications. Intuitively, since optimal beam selection can benefit from the knowledge of the positions of communication terminals, there has been increasing interest in leveraging position data to reduce the overhead in mmWave beam prediction. Prior work, however, studied this problem using only synthetic data that generally does not accurately represent real-world measurements. In this paper, we investigate position-aided beam prediction using a real-world large-scale dataset to derive insights into precisely how much overhead can be saved in practice. Furthermore, we analyze which machine learning algorithms perform best, what factors degrade inference performance in real data, and which machine learning metrics are more meaningful in capturing the actual communication system performance.

arXiv:2205.07831공식 코드 · Papers with Code

Expected Frequency Matrices of Elections: Computation, Geometry, and Preference Learning

We use the ``map of elections'' approach of Szufa et al. (AAMAS-2020) to analyze several well-known vote distributions. For each of them, we give an explicit formula or an efficient algorithm for computing its frequency matrix, which captures the probability that a given candidate appears in a given position in a sampled vote. We use these matrices to draw the ``skeleton map'' of distributions, evaluate its robustness, and analyze its properties. Finally, we develop a general and unified framework for learning the distribution of real-world preferences using the frequency matrices of established vote distributions.

arXiv:2205.07334공식 코드 · Papers with Code

Mack-Net model: Blending Mack's model with Recurrent Neural Networks

In general insurance companies, a correct estimation of liabilities plays a key role due to its impact on management and investing decisions. Since the Financial Crisis of 2007-2008 and the strengthening of regulation, the focus is not only on the total reserve but also on its variability, which is an indicator of the risk assumed by the company. Thus, measures that relate profitability with risk are crucial in order to understand the financial position of insurance firms. Taking advantage of the increasing computational power, this paper introduces a stochastic reserving model whose aim is to improve the performance of the traditional Mack's reserving model by applying an ensemble of Recurrent Neural Networks. The results demonstrate that blending traditional reserving models with deep and machine learning techniques leads to a more accurate assessment of general insurance liabilities.

arXiv:2205.07177공식 코드 · Papers with Code

Hero-Gang Neural Model For Named Entity Recognition

Named entity recognition (NER) is a fundamental and important task in NLP, aiming at identifying named entities (NEs) from free text. Recently, since the multi-head attention mechanism applied in the Transformer model can effectively capture longer contextual information, Transformer-based models have become the mainstream methods and have achieved significant performance in this task. Unfortunately, although these models can capture effective global context information, they are still limited in the local feature and position information extraction, which is critical in NER. In this paper, to address this limitation, we propose a novel Hero-Gang Neural structure (HGN), including the Hero and Gang module, to leverage both global and local information to promote NER. Specifically, the Hero module is composed of a Transformer-based encoder to maintain the advantage of the self-attention mechanism, and the Gang module utilizes a multi-window recurrent module to extract local features and position information under the guidance of the Hero module. Afterward, the proposed multi-window attention effectively combines global information and multiple local features for predicting entity labels. Experimental results on several benchmark datasets demonstrate the effectiveness of our proposed model.

arXiv:2205.05614공식 코드 · Papers with Code

Gamma and Vega Hedging Using Deep Distributional Reinforcement Learning

We show how D4PG can be used in conjunction with quantile regression to develop a hedging strategy for a trader responsible for derivatives that arrive stochastically and depend on a single underlying asset. We assume that the trader makes the portfolio delta neutral at the end of each day by taking a position in the underlying asset. We focus on how trades in the options can be used to manage gamma and vega. The option trades are subject to transaction costs. We consider three different objective functions. We reach conclusions on how the optimal hedging strategy depends on the trader's objective function, the level of transaction costs, and the maturity of the options used for hedging. We also investigate the robustness of the hedging strategy to the process assumed for the underlying asset.

arXiv:2205.05600공식 코드 · Papers with Code

RLOP: RL Methods in Option Pricing from a Mathematical Perspective

Abstract In this work, we build two environments, namely the modified QLBS and RLOP models, from a mathematics perspective which enables RL methods in option pricing through replicating by portfolio. We implement the environment specifications (the source code can be found at https://github.com/owen8877/RLOP), the learning algorithm, and agent parametrization by a neural network. The learned optimal hedging strategy is compared against the BS prediction. The effect of various factors is considered and studied based on how they affect the optimal price and position.

arXiv:2205.03972공식 코드 · Papers with Code

Robust (Controlled) Table-to-Text Generation with Structure-Aware Equivariance Learning

Controlled table-to-text generation seeks to generate natural language descriptions for highlighted subparts of a table. Previous SOTA systems still employ a sequence-to-sequence generation method, which merely captures the table as a linear structure and is brittle when table layouts change. We seek to go beyond this paradigm by (1) effectively expressing the relations of content pieces in the table, and (2) making our model robust to content-invariant structural transformations. Accordingly, we propose an equivariance learning framework, which encodes tables with a structure-aware self-attention mechanism. This prunes the full self-attention structure into an order-invariant graph attention that captures the connected graph structure of cells belonging to the same row or column, and it differentiates between relevant cells and irrelevant cells from the structural perspective. Our framework also modifies the positional encoding mechanism to preserve the relative position of tokens in the same cell but enforce position invariance among different cells. Our technology is free to be plugged into existing table-to-text generation models, and has improved T5-based models to offer better performance on ToTTo and HiTab. Moreover, on a harder version of ToTTo, we preserve promising performance, while previous SOTA systems, even with transformation-based data augmentation, have seen significant performance drops. Our code is available at https://github.com/luka-group/Lattice.

arXiv:2205.03212공식 코드 · Papers with Code

Predicting Future Occupancy Grids in Dynamic Environment with Spatio-Temporal Learning

Reliably predicting future occupancy of highly dynamic urban environments is an important precursor for safe autonomous navigation. Common challenges in the prediction include forecasting the relative position of other vehicles, modelling the dynamics of vehicles subjected to different traffic conditions, and vanishing surrounding objects. To tackle these challenges, we propose a spatio-temporal prediction network pipeline that takes the past information from the environment and semantic labels separately for generating future occupancy predictions. Compared to the current SOTA, our approach predicts occupancy for a longer horizon of 3 seconds and in a relatively complex environment from the nuScenes dataset. Our experimental results demonstrate the ability of spatio-temporal networks to understand scene dynamics without the need for HD-Maps and explicit modeling dynamic objects. We publicly release our occupancy grid dataset based on nuScenes to support further research.

arXiv:2204.12807공식 코드 · Papers with Code

Probing Simile Knowledge from Pre-trained Language Models

Simile interpretation (SI) and simile generation (SG) are challenging tasks for NLP because models require adequate world knowledge to produce predictions. Previous works have employed many hand-crafted resources to bring knowledge-related into models, which is time-consuming and labor-intensive. In recent years, pre-trained language models (PLMs) based approaches have become the de-facto standard in NLP since they learn generic knowledge from a large corpus. The knowledge embedded in PLMs may be useful for SI and SG tasks. Nevertheless, there are few works to explore it. In this paper, we probe simile knowledge from PLMs to solve the SI and SG tasks in the unified framework of simile triple completion for the first time. The backbone of our framework is to construct masked sentences with manual patterns and then predict the candidate words in the masked position. In this framework, we adopt a secondary training process (Adjective-Noun mask Training) with the masked language model (MLM) loss to enhance the prediction diversity of candidate words in the masked position. Moreover, pattern ensemble (PE) and pattern search (PS) are applied to improve the quality of predicted words. Finally, automatic and human evaluations demonstrate the effectiveness of our framework in both SI and SG tasks.

arXiv:2204.05891공식 코드 · Papers with Code

A DNN Framework for Learning Lagrangian Drift With Uncertainty

Reconstructions of Lagrangian drift, for example for objects lost at sea, are often uncertain due to unresolved physical phenomena within the data. Uncertainty is usually overcome by introducing stochasticity into the drift, but this approach requires specific assumptions for modelling uncertainty. We remove this constraint by presenting a purely data-driven framework for modelling probabilistic drift in flexible environments. Using ocean circulation model simulations, we generate probabilistic trajectories of object location by simulating uncertainty in the initial object position. We train an emulator of probabilistic drift over one day given perfectly known velocities and observe good agreement with numerical simulations. Several loss functions are tested. Then, we strain our framework by training models where the input information is imperfect. On these harder scenarios, we observe reasonable predictions although the effects of data drift become noticeable when evaluating the models against unseen flow scenarios.

arXiv:2204.04534공식 코드 · Papers with Code

Extending the Scope of Out-of-Domain: Examining QA models in multiple subdomains

Past works that investigate out-of-domain performance of QA systems have mainly focused on general domains (e.g. news domain, wikipedia domain), underestimating the importance of subdomains defined by the internal characteristics of QA datasets. In this paper, we extend the scope of "out-of-domain" by splitting QA examples into different subdomains according to their several internal characteristics including question type, text length, answer position. We then examine the performance of QA systems trained on the data from different subdomains. Experimental results show that the performance of QA systems can be significantly reduced when the train data and test data come from different subdomains. These results question the generalizability of current QA systems in multiple subdomains, suggesting the need to combat the bias introduced by the internal characteristics of QA datasets.

arXiv:2204.02637공식 코드 · Papers with Code

Global HRTF Interpolation via Learned Affine Transformation of Hyper-conditioned Features

Estimating Head-Related Transfer Functions (HRTFs) of arbitrary source points is essential in immersive binaural audio rendering. Computing each individual's HRTFs is challenging, as traditional approaches require expensive time and computational resources, while modern data-driven approaches are data-hungry. Especially for the data-driven approaches, existing HRTF datasets differ in spatial sampling distributions of source positions, posing a major problem when generalizing the method across multiple datasets. To alleviate this, we propose a deep learning method based on a novel conditioning architecture. The proposed method can predict an HRTF of any position by interpolating the HRTFs of known distributions. Experimental results show that the proposed architecture improves the model's generalizability across datasets with various coordinate systems. Additional demonstrations show that the model robustly reconstructs the target HRTFs from the spatially downsampled HRTFs in both quantitative and perceptual measures.

arXiv:2204.02119공식 코드 · Papers with Code

Transition Information Enhanced Disentangled Graph Neural Networks for Session-based Recommendation

Session-based recommendation is a practical recommendation task that predicts the next item based on an anonymous behavior sequence, and its performance relies heavily on the transition information between items in the sequence. The SOTA methods in SBR employ GNN to model neighboring item transitions from global (i.e, other sessions) and local (i.e, current session) contexts. However, most existing methods treat neighbors from different sessions equally without considering that the neighbor items from different sessions may share similar features with the target item on different aspects and may have different contributions. In other words, they have not explored finer-granularity transition information between items in the global context, leading to sub-optimal performance. In this paper, we fill this gap by proposing a novel Transition Information Enhanced Disentangled Graph Neural Network (TIE-DGNN) model to capture finer-granular transition information between items and try to interpret the reason of the transition by modeling the various factors of the item. Specifically, we propose a position-aware global graph, which utilizes the relative position information to model the neighboring item transition. Then, we slice item embeddings into blocks, each of which represents a factor, and use disentangling module to separately learn the factor embeddings over the global graph. For local context, we train item embeddings by using attention mechanisms to capture transition information from the current session. To this end, our model considers two levels of transition information. Especially in global text, we not only consider finer-granularity transition information between items but also take user intents at factor-level into account to interpret the key reason for the transition. Extensive experiments on three datasets demonstrate the superiority of our method over the SOTA methods.

arXiv:2204.02179공식 코드 · Papers with Code

Towards Robust and Accurate Myoelectric Controller Design based on Multi-objective Optimization using Evolutionary Computation

Myoelectric pattern recognition is one of the important aspects in the design of the control strategy for various applications including upper-limb prostheses and bio-robotic hand movement systems. The current work has proposed an approach to design an energy-efficient EMG-based controller by considering a kernelized SVM classifier for decoding the information of surface electromyography (sEMG) signals to infer the underlying muscle movements. In order to achieve the optimized performance of the EMG-based controller, our main strategy of classifier design is to reduce the false movements of the overall system (when the EMG-based controller is at the `Rest' position). To this end, we have formulated the training algorithm of the proposed supervised learning system as a general constrained multi-objective optimization problem. An elitist multi-objective evolutionary algorithm $-$ the non-dominated sorting genetic algorithm II (NSGA-II) has been used to tune the hyperparameters of SVM. We have presented the experimental results by performing the experiments on a dataset consisting of the sEMG signals collected from eleven subjects at five different upper limb positions. Furthermore, the performance of the trained models based on the two-objective metrics, namely classification accuracy, and false-negative have been evaluated on two different test sets to examine the generalization capability of the proposed training approach while implementing limb-position invariant EMG classification. It is evident from the presented result that the proposed approach provides much more flexibility to the designer in selecting the parameters of the classifier to optimize the energy efficiency of the EMG-based controller.

arXiv:2204.00815공식 코드 · Papers with Code

Unbiased Top-k Learning to Rank with Causal Likelihood Decomposition

Unbiased learning to rank has been proposed to alleviate the biases in the search ranking, making it possible to train ranking models with user interaction data. In real applications, search engines are designed to display only the most relevant k documents from the retrieved candidate set. The rest candidates are discarded. As a consequence, position bias and sample selection bias usually occur simultaneously. Existing unbiased learning to rank approaches either focus on one type of bias (e.g., position bias) or mitigate the position bias and sample selection bias with separate components, overlooking their associations. In this study, we first analyze the mechanisms and associations of position bias and sample selection bias from the viewpoint of a causal graph. Based on the analysis, we propose Causal Likelihood Decomposition (CLD), a unified approach to simultaneously mitigating these two biases in top-k learning to rank. By decomposing the log-likelihood of the biased data as an unbiased term that only related to relevance, plus other terms related to biases, CLD successfully detaches the relevance from position bias and sample selection bias. An unbiased ranking model can be obtained from the unbiased term, via maximizing the whole likelihood. An extension to the pairwise neural ranking is also developed. Advantages of CLD include theoretical soundness and a unified framework for pointwise and pairwise unbiased top-k learning to rank. Extensive experimental results verified that CLD, including its pairwise neural extension, outperformed the baselines by mitigating both the position bias and the sample selection bias. Empirical studies also showed that CLD is robust to the variation of bias severity and the click noise.

arXiv:2204.00718공식 코드 · Papers with Code

Implicit Feedback for Dense Passage Retrieval: A Counterfactual Approach

In this paper we study how to effectively exploit implicit feedback in Dense Retrievers (DRs). We consider the specific case in which click data from a historic click log is available as implicit feedback. We then exploit such historic implicit interactions to improve the effectiveness of a DR. A key challenge that we study is the effect that biases in the click signal, such as position bias, have on the DRs. To overcome the problems associated with the presence of such bias, we propose the Counterfactual Rocchio (CoRocchio) algorithm for exploiting implicit feedback in Dense Retrievers. We demonstrate both theoretically and empirically that dense query representations learnt with CoRocchio are unbiased with respect to position bias and lead to higher retrieval effectiveness. We make available the implementations of the proposed methods and the experimental framework, along with all results at https://github.com/ielab/Counterfactual-DR.

arXiv:2203.17118공식 코드 · Papers with Code

Doubly-Robust Estimation for Correcting Position-Bias in Click Feedback for Unbiased Learning to Rank

Clicks on rankings suffer from position-bias: generally items on lower ranks are less likely to be examined - and thus clicked - by users, in spite of their actual preferences between items. The prevalent approach to unbiased click-based learning-to-rank (LTR) is based on counterfactual inverse-propensity-scoring (IPS) estimation. In contrast with general reinforcement learning, counterfactual doubly-robust (DR) estimation has not been applied to click-based LTR in previous literature. In this paper, we introduce a novel DR estimator that is the first DR approach specifically designed for position-bias. The difficulty with position-bias is that the treatment - user examination - is not directly observable in click data. As a solution, our estimator uses the expected treatment per rank, instead of the actual treatment that existing DR estimators use. Our novel DR estimator has more robust unbiasedness conditions than the existing IPS approach, and in addition, provides enormous decreases in variance: our experimental results indicate it requires several orders of magnitude fewer datapoints to converge at optimal performance. For the unbiased LTR field, our DR estimator contributes both increases in state-of-the-art performance and the most robust theoretical guarantees of all known LTR estimators.

arXiv:2203.16634공식 코드 · Papers with Code

Transformer Language Models without Positional Encodings Still Learn Positional Information

Causal transformer language models (LMs), such as GPT-3, typically require some form of positional encoding, such as positional embeddings. However, we show that LMs without any explicit positional encoding are still competitive with standard models, and that this phenomenon is robust across different datasets, model sizes, and sequence lengths. Probing experiments reveal that such models acquire an implicit notion of absolute positions throughout the network, effectively compensating for the missing information. We conjecture that causal attention enables the model to infer the number of predecessors that each token can attend to, thereby approximating its absolute position. Our findings indicate that causal LMs might derive positional awareness not only from the explicit positioning mechanism, but also from the effects of the causal mask.