Pub Date : 2023-04-20DOI: 10.48550/arXiv.2304.10668
Costas Mavromatis, V. N. Ioannidis, Shen Wang, Da Zheng, Soji Adeshina, Jun Ma, Han Zhao, C. Faloutsos, G. Karypis
How can we learn effective node representations on textual graphs? Graph Neural Networks (GNNs) that use Language Models (LMs) to encode textual information of graphs achieve state-of-the-art performance in many node classification tasks. Yet, combining GNNs with LMs has not been widely explored for practical deployments due to its scalability issues. In this work, we tackle this challenge by developing a Graph-Aware Distillation framework (GRAD) to encode graph structures into an LM for graph-free, fast inference. Different from conventional knowledge distillation, GRAD jointly optimizes a GNN teacher and a graph-free student over the graph's nodes via a shared LM. This encourages the graph-free student to exploit graph information encoded by the GNN teacher while at the same time, enables the GNN teacher to better leverage textual information from unlabeled nodes. As a result, the teacher and the student models learn from each other to improve their overall performance. Experiments in eight node classification benchmarks in both transductive and inductive settings showcase GRAD's superiority over existing distillation approaches for textual graphs.
{"title":"Train Your Own GNN Teacher: Graph-Aware Distillation on Textual Graphs","authors":"Costas Mavromatis, V. N. Ioannidis, Shen Wang, Da Zheng, Soji Adeshina, Jun Ma, Han Zhao, C. Faloutsos, G. Karypis","doi":"10.48550/arXiv.2304.10668","DOIUrl":"https://doi.org/10.48550/arXiv.2304.10668","url":null,"abstract":"How can we learn effective node representations on textual graphs? Graph Neural Networks (GNNs) that use Language Models (LMs) to encode textual information of graphs achieve state-of-the-art performance in many node classification tasks. Yet, combining GNNs with LMs has not been widely explored for practical deployments due to its scalability issues. In this work, we tackle this challenge by developing a Graph-Aware Distillation framework (GRAD) to encode graph structures into an LM for graph-free, fast inference. Different from conventional knowledge distillation, GRAD jointly optimizes a GNN teacher and a graph-free student over the graph's nodes via a shared LM. This encourages the graph-free student to exploit graph information encoded by the GNN teacher while at the same time, enables the GNN teacher to better leverage textual information from unlabeled nodes. As a result, the teacher and the student models learn from each other to improve their overall performance. Experiments in eight node classification benchmarks in both transductive and inductive settings showcase GRAD's superiority over existing distillation approaches for textual graphs.","PeriodicalId":74091,"journal":{"name":"Machine learning and knowledge discovery in databases : European Conference, ECML PKDD ... : proceedings. ECML PKDD (Conference)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-04-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79203997","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-04-18DOI: 10.48550/arXiv.2304.08841
Bosong Huang, Weihao Yu, Ruzhong Xie, Jing Xiao, Jin Huang
Source localization is the inverse problem of graph information dissemination and has broad practical applications. However, the inherent intricacy and uncertainty in information dissemination pose significant challenges, and the ill-posed nature of the source localization problem further exacerbates these challenges. Recently, deep generative models, particularly diffusion models inspired by classical non-equilibrium thermodynamics, have made significant progress. While diffusion models have proven to be powerful in solving inverse problems and producing high-quality reconstructions, applying them directly to the source localization is infeasible for two reasons. Firstly, it is impossible to calculate the posterior disseminated results on a large-scale network for iterative denoising sampling, which would incur enormous computational costs. Secondly, in the existing methods for this field, the training data itself are ill-posed (many-to-one); thus simply transferring the diffusion model would only lead to local optima. To address these challenges, we propose a two-stage optimization framework, the source localization denoising diffusion model (SL-Diff). In the coarse stage, we devise the source proximity degrees as the supervised signals to generate coarse-grained source predictions. This aims to efficiently initialize the next stage, significantly reducing its convergence time and calibrating the convergence process. Furthermore, the introduction of cascade temporal information in this training method transforms the many-to-one mapping relationship into a one-to-one relationship, perfectly addressing the ill-posed problem. In the fine stage, we design a diffusion model for the graph inverse problem that can quantify the uncertainty in the dissemination. The proposed SL-Diff yields excellent prediction results within a reasonable sampling time at extensive experiments.
{"title":"Two-stage Denoising Diffusion Model for Source Localization in Graph Inverse Problems","authors":"Bosong Huang, Weihao Yu, Ruzhong Xie, Jing Xiao, Jin Huang","doi":"10.48550/arXiv.2304.08841","DOIUrl":"https://doi.org/10.48550/arXiv.2304.08841","url":null,"abstract":"Source localization is the inverse problem of graph information dissemination and has broad practical applications. However, the inherent intricacy and uncertainty in information dissemination pose significant challenges, and the ill-posed nature of the source localization problem further exacerbates these challenges. Recently, deep generative models, particularly diffusion models inspired by classical non-equilibrium thermodynamics, have made significant progress. While diffusion models have proven to be powerful in solving inverse problems and producing high-quality reconstructions, applying them directly to the source localization is infeasible for two reasons. Firstly, it is impossible to calculate the posterior disseminated results on a large-scale network for iterative denoising sampling, which would incur enormous computational costs. Secondly, in the existing methods for this field, the training data itself are ill-posed (many-to-one); thus simply transferring the diffusion model would only lead to local optima. To address these challenges, we propose a two-stage optimization framework, the source localization denoising diffusion model (SL-Diff). In the coarse stage, we devise the source proximity degrees as the supervised signals to generate coarse-grained source predictions. This aims to efficiently initialize the next stage, significantly reducing its convergence time and calibrating the convergence process. Furthermore, the introduction of cascade temporal information in this training method transforms the many-to-one mapping relationship into a one-to-one relationship, perfectly addressing the ill-posed problem. In the fine stage, we design a diffusion model for the graph inverse problem that can quantify the uncertainty in the dissemination. The proposed SL-Diff yields excellent prediction results within a reasonable sampling time at extensive experiments.","PeriodicalId":74091,"journal":{"name":"Machine learning and knowledge discovery in databases : European Conference, ECML PKDD ... : proceedings. ECML PKDD (Conference)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-04-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83696207","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-04-17DOI: 10.48550/arXiv.2304.08453
Victor Agostinelli, Lizhong Chen
Various natural language processing (NLP) tasks necessitate models that are efficient and small based on their ultimate application at the edge or in other resource-constrained environments. While prior research has reduced the size of these models, increasing computational efficiency without considerable performance impacts remains difficult, especially for autoregressive tasks. This paper proposes modular linearized attention (MLA), which combines multiple efficient attention mechanisms, including cosFormer, to maximize inference quality while achieving notable speedups. We validate this approach on several autoregressive NLP tasks, including speech-to-text neural machine translation (S2T NMT), speech-to-text simultaneous translation (SimulST), and autoregressive text-to-spectrogram, noting efficiency gains on TTS and competitive performance for NMT and SimulST during training and inference.
{"title":"Improving Autoregressive NLP Tasks via Modular Linearized Attention","authors":"Victor Agostinelli, Lizhong Chen","doi":"10.48550/arXiv.2304.08453","DOIUrl":"https://doi.org/10.48550/arXiv.2304.08453","url":null,"abstract":"Various natural language processing (NLP) tasks necessitate models that are efficient and small based on their ultimate application at the edge or in other resource-constrained environments. While prior research has reduced the size of these models, increasing computational efficiency without considerable performance impacts remains difficult, especially for autoregressive tasks. This paper proposes modular linearized attention (MLA), which combines multiple efficient attention mechanisms, including cosFormer, to maximize inference quality while achieving notable speedups. We validate this approach on several autoregressive NLP tasks, including speech-to-text neural machine translation (S2T NMT), speech-to-text simultaneous translation (SimulST), and autoregressive text-to-spectrogram, noting efficiency gains on TTS and competitive performance for NMT and SimulST during training and inference.","PeriodicalId":74091,"journal":{"name":"Machine learning and knowledge discovery in databases : European Conference, ECML PKDD ... : proceedings. ECML PKDD (Conference)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88071814","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-04-17DOI: 10.48550/arXiv.2304.08349
Rishi Hazra, L. D. Raedt
Despite numerous successes in Deep Reinforcement Learning (DRL), the learned policies are not interpretable. Moreover, since DRL does not exploit symbolic relational representations, it has difficulties in coping with structural changes in its environment (such as increasing the number of objects). Relational Reinforcement Learning, on the other hand, inherits the relational representations from symbolic planning to learn reusable policies. However, it has so far been unable to scale up and exploit the power of deep neural networks. We propose Deep Explainable Relational Reinforcement Learning (DERRL), a framework that exploits the best of both -- neural and symbolic worlds. By resorting to a neuro-symbolic approach, DERRL combines relational representations and constraints from symbolic planning with deep learning to extract interpretable policies. These policies are in the form of logical rules that explain how each decision (or action) is arrived at. Through several experiments, in setups like the Countdown Game, Blocks World, Gridworld, and Traffic, we show that the policies learned by DERRL can be applied to different configurations and contexts, hence generalizing to environmental modifications.
{"title":"Deep Explainable Relational Reinforcement Learning: A Neuro-Symbolic Approach","authors":"Rishi Hazra, L. D. Raedt","doi":"10.48550/arXiv.2304.08349","DOIUrl":"https://doi.org/10.48550/arXiv.2304.08349","url":null,"abstract":"Despite numerous successes in Deep Reinforcement Learning (DRL), the learned policies are not interpretable. Moreover, since DRL does not exploit symbolic relational representations, it has difficulties in coping with structural changes in its environment (such as increasing the number of objects). Relational Reinforcement Learning, on the other hand, inherits the relational representations from symbolic planning to learn reusable policies. However, it has so far been unable to scale up and exploit the power of deep neural networks. We propose Deep Explainable Relational Reinforcement Learning (DERRL), a framework that exploits the best of both -- neural and symbolic worlds. By resorting to a neuro-symbolic approach, DERRL combines relational representations and constraints from symbolic planning with deep learning to extract interpretable policies. These policies are in the form of logical rules that explain how each decision (or action) is arrived at. Through several experiments, in setups like the Countdown Game, Blocks World, Gridworld, and Traffic, we show that the policies learned by DERRL can be applied to different configurations and contexts, hence generalizing to environmental modifications.","PeriodicalId":74091,"journal":{"name":"Machine learning and knowledge discovery in databases : European Conference, ECML PKDD ... : proceedings. ECML PKDD (Conference)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86261815","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-04-03DOI: 10.48550/arXiv.2304.01169
Kai Ouyang, Wenhao Zheng, Chen Tang, Xuanji Xiao, Haitao Zheng
Post-click Conversion Rate (CVR) prediction task plays an essential role in industrial applications, such as recommendation and advertising. Conventional CVR methods typically suffer from the data sparsity problem as they rely only on samples where the user has clicked. To address this problem, researchers have introduced the method of multi-task learning, which utilizes non-clicked samples and shares feature representations of the Click-Through Rate (CTR) task with the CVR task. However, it should be noted that the CVR and CTR tasks are fundamentally different and may even be contradictory. Therefore, introducing a large amount of CTR information without distinction may drown out valuable information related to CVR. This phenomenon is called the curse of knowledge problem in this paper. To tackle this issue, we argue that a trade-off should be achieved between the introduction of large amounts of auxiliary information and the protection of valuable information related to CVR. Hence, we propose a Click-aware Structure Transfer model with sample Weight Assignment, abbreviated as CSTWA. It pays more attention to the latent structure information, which can filter the input information that is related to CVR, instead of directly sharing feature representations. Meanwhile, to capture the representation conflict between CTR and CVR, we calibrate the representation layer and reweight the discriminant layer to excavate the click bias information from the CTR tower. Moreover, it incorporates a sample weight assignment algorithm biased towards CVR modeling, to make the knowledge from CTR would not mislead the CVR. Extensive experiments on industrial and public datasets have demonstrated that CSTWA significantly outperforms widely used and competitive models.
{"title":"Click-aware Structure Transfer with Sample Weight Assignment for Post-Click Conversion Rate Estimation","authors":"Kai Ouyang, Wenhao Zheng, Chen Tang, Xuanji Xiao, Haitao Zheng","doi":"10.48550/arXiv.2304.01169","DOIUrl":"https://doi.org/10.48550/arXiv.2304.01169","url":null,"abstract":"Post-click Conversion Rate (CVR) prediction task plays an essential role in industrial applications, such as recommendation and advertising. Conventional CVR methods typically suffer from the data sparsity problem as they rely only on samples where the user has clicked. To address this problem, researchers have introduced the method of multi-task learning, which utilizes non-clicked samples and shares feature representations of the Click-Through Rate (CTR) task with the CVR task. However, it should be noted that the CVR and CTR tasks are fundamentally different and may even be contradictory. Therefore, introducing a large amount of CTR information without distinction may drown out valuable information related to CVR. This phenomenon is called the curse of knowledge problem in this paper. To tackle this issue, we argue that a trade-off should be achieved between the introduction of large amounts of auxiliary information and the protection of valuable information related to CVR. Hence, we propose a Click-aware Structure Transfer model with sample Weight Assignment, abbreviated as CSTWA. It pays more attention to the latent structure information, which can filter the input information that is related to CVR, instead of directly sharing feature representations. Meanwhile, to capture the representation conflict between CTR and CVR, we calibrate the representation layer and reweight the discriminant layer to excavate the click bias information from the CTR tower. Moreover, it incorporates a sample weight assignment algorithm biased towards CVR modeling, to make the knowledge from CTR would not mislead the CVR. Extensive experiments on industrial and public datasets have demonstrated that CSTWA significantly outperforms widely used and competitive models.","PeriodicalId":74091,"journal":{"name":"Machine learning and knowledge discovery in databases : European Conference, ECML PKDD ... : proceedings. ECML PKDD (Conference)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-04-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76983112","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Temporal knowledge graph completion (TKGC) aims to predict the missing links among the entities in a temporal knwoledge graph (TKG). Most previous TKGC methods only consider predicting the missing links among the entities seen in the training set, while they are unable to achieve great performance in link prediction concerning newly-emerged unseen entities. Recently, a new task, i.e., TKG few-shot out-of-graph (OOG) link prediction, is proposed, where TKGC models are required to achieve great link prediction performance concerning newly-emerged entities that only have few-shot observed examples. In this work, we propose a TKGC method FITCARL that combines few-shot learning with reinforcement learning to solve this task. In FITCARL, an agent traverses through the whole TKG to search for the prediction answer. A policy network is designed to guide the search process based on the traversed path. To better address the data scarcity problem in the few-shot setting, we introduce a module that computes the confidence of each candidate action and integrate it into the policy for action selection. We also exploit the entity concept information with a novel concept regularizer to boost model performance. Experimental results show that FITCARL achieves stat-of-the-art performance on TKG few-shot OOG link prediction.
{"title":"Improving Few-Shot Inductive Learning on Temporal Knowledge Graphs using Confidence-Augmented Reinforcement Learning","authors":"Zifeng Ding, Jingpei Wu, Zong-Xun Li, Yunpu Ma, Volker Tresp","doi":"10.48550/arXiv.2304.00613","DOIUrl":"https://doi.org/10.48550/arXiv.2304.00613","url":null,"abstract":"Temporal knowledge graph completion (TKGC) aims to predict the missing links among the entities in a temporal knwoledge graph (TKG). Most previous TKGC methods only consider predicting the missing links among the entities seen in the training set, while they are unable to achieve great performance in link prediction concerning newly-emerged unseen entities. Recently, a new task, i.e., TKG few-shot out-of-graph (OOG) link prediction, is proposed, where TKGC models are required to achieve great link prediction performance concerning newly-emerged entities that only have few-shot observed examples. In this work, we propose a TKGC method FITCARL that combines few-shot learning with reinforcement learning to solve this task. In FITCARL, an agent traverses through the whole TKG to search for the prediction answer. A policy network is designed to guide the search process based on the traversed path. To better address the data scarcity problem in the few-shot setting, we introduce a module that computes the confidence of each candidate action and integrate it into the policy for action selection. We also exploit the entity concept information with a novel concept regularizer to boost model performance. Experimental results show that FITCARL achieves stat-of-the-art performance on TKG few-shot OOG link prediction.","PeriodicalId":74091,"journal":{"name":"Machine learning and knowledge discovery in databases : European Conference, ECML PKDD ... : proceedings. ECML PKDD (Conference)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-04-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73925491","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-03-24DOI: 10.48550/arXiv.2303.13808
K. Mehta, Anuj Mahajan, Priyesh Kumar
Recent advances in Reinforcement Learning (RL) have led to many exciting applications. These advancements have been driven by improvements in both algorithms and engineering, which have resulted in faster training of RL agents. We present marl-jax, a multi-agent reinforcement learning software package for training and evaluating social generalization of the agents. The package is designed for training a population of agents in multi-agent environments and evaluating their ability to generalize to diverse background agents. It is built on top of DeepMind's JAX ecosystem~cite{deepmind2020jax} and leverages the RL ecosystem developed by DeepMind. Our framework marl-jax is capable of working in cooperative and competitive, simultaneous-acting environments with multiple agents. The package offers an intuitive and user-friendly command-line interface for training a population and evaluating its generalization capabilities. In conclusion, marl-jax provides a valuable resource for researchers interested in exploring social generalization in the context of MARL. The open-source code for marl-jax is available at: href{https://github.com/kinalmehta/marl-jax}{https://github.com/kinalmehta/marl-jax}
{"title":"marl-jax: Multi-agent Reinforcement Leaning framework for Social Generalization","authors":"K. Mehta, Anuj Mahajan, Priyesh Kumar","doi":"10.48550/arXiv.2303.13808","DOIUrl":"https://doi.org/10.48550/arXiv.2303.13808","url":null,"abstract":"Recent advances in Reinforcement Learning (RL) have led to many exciting applications. These advancements have been driven by improvements in both algorithms and engineering, which have resulted in faster training of RL agents. We present marl-jax, a multi-agent reinforcement learning software package for training and evaluating social generalization of the agents. The package is designed for training a population of agents in multi-agent environments and evaluating their ability to generalize to diverse background agents. It is built on top of DeepMind's JAX ecosystem~cite{deepmind2020jax} and leverages the RL ecosystem developed by DeepMind. Our framework marl-jax is capable of working in cooperative and competitive, simultaneous-acting environments with multiple agents. The package offers an intuitive and user-friendly command-line interface for training a population and evaluating its generalization capabilities. In conclusion, marl-jax provides a valuable resource for researchers interested in exploring social generalization in the context of MARL. The open-source code for marl-jax is available at: href{https://github.com/kinalmehta/marl-jax}{https://github.com/kinalmehta/marl-jax}","PeriodicalId":74091,"journal":{"name":"Machine learning and knowledge discovery in databases : European Conference, ECML PKDD ... : proceedings. ECML PKDD (Conference)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-03-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90815754","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-03-09DOI: 10.48550/arXiv.2303.05018
Junfan Li, Shizhong Liao
In this paper, we improve the regret bound for online kernel selection under bandit feedback. Previous algorithm enjoys a $O((Vert fVert^2_{mathcal{H}_i}+1)K^{frac{1}{3}}T^{frac{2}{3}})$ expected bound for Lipschitz loss functions. We prove two types of regret bounds improving the previous bound. For smooth loss functions, we propose an algorithm with a $O(U^{frac{2}{3}}K^{-frac{1}{3}}(sum^K_{i=1}L_T(f^ast_i))^{frac{2}{3}})$ expected bound where $L_T(f^ast_i)$ is the cumulative losses of optimal hypothesis in $mathbb{H}_{i}={finmathcal{H}_i:Vert fVert_{mathcal{H}_i}leq U}$. The data-dependent bound keeps the previous worst-case bound and is smaller if most of candidate kernels match well with the data. For Lipschitz loss functions, we propose an algorithm with a $O(Usqrt{KT}ln^{frac{2}{3}}{T})$ expected bound asymptotically improving the previous bound. We apply the two algorithms to online kernel selection with time constraint and prove new regret bounds matching or improving the previous $O(sqrt{Tln{K}} +Vert fVert^2_{mathcal{H}_i}max{sqrt{T},frac{T}{sqrt{mathcal{R}}}})$ expected bound where $mathcal{R}$ is the time budget. Finally, we empirically verify our algorithms on online regression and classification tasks.
{"title":"Improved Regret Bounds for Online Kernel Selection under Bandit Feedback","authors":"Junfan Li, Shizhong Liao","doi":"10.48550/arXiv.2303.05018","DOIUrl":"https://doi.org/10.48550/arXiv.2303.05018","url":null,"abstract":"In this paper, we improve the regret bound for online kernel selection under bandit feedback. Previous algorithm enjoys a $O((Vert fVert^2_{mathcal{H}_i}+1)K^{frac{1}{3}}T^{frac{2}{3}})$ expected bound for Lipschitz loss functions. We prove two types of regret bounds improving the previous bound. For smooth loss functions, we propose an algorithm with a $O(U^{frac{2}{3}}K^{-frac{1}{3}}(sum^K_{i=1}L_T(f^ast_i))^{frac{2}{3}})$ expected bound where $L_T(f^ast_i)$ is the cumulative losses of optimal hypothesis in $mathbb{H}_{i}={finmathcal{H}_i:Vert fVert_{mathcal{H}_i}leq U}$. The data-dependent bound keeps the previous worst-case bound and is smaller if most of candidate kernels match well with the data. For Lipschitz loss functions, we propose an algorithm with a $O(Usqrt{KT}ln^{frac{2}{3}}{T})$ expected bound asymptotically improving the previous bound. We apply the two algorithms to online kernel selection with time constraint and prove new regret bounds matching or improving the previous $O(sqrt{Tln{K}} +Vert fVert^2_{mathcal{H}_i}max{sqrt{T},frac{T}{sqrt{mathcal{R}}}})$ expected bound where $mathcal{R}$ is the time budget. Finally, we empirically verify our algorithms on online regression and classification tasks.","PeriodicalId":74091,"journal":{"name":"Machine learning and knowledge discovery in databases : European Conference, ECML PKDD ... : proceedings. ECML PKDD (Conference)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"91255541","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-02-27DOI: 10.48550/arXiv.2302.13563
Junwoo Park, Jungsoo Lee, Youngin Cho, W. Shin, Dongmin Kim, J. Choo, E. Choi
Time-series forecasting models often encounter abrupt changes in a given period of time which generally occur due to unexpected or unknown events. Despite their scarce occurrences in the training set, abrupt changes incur loss that significantly contributes to the total loss. Therefore, they act as noisy training samples and prevent the model from learning generalizable patterns, namely the normal states. Based on our findings, we propose a reweighting framework that down-weights the losses incurred by abrupt changes and up-weights those by normal states. For the reweighting framework, we first define a measurement termed Local Discrepancy (LD) which measures the degree of abruptness of a change in a given period of time. Since a training set is mostly composed of normal states, we then consider how frequently the temporal changes appear in the training set based on LD. Our reweighting framework is applicable to existing time-series forecasting models regardless of the architectures. Through extensive experiments on 12 time-series forecasting models over eight datasets with various in-output sequence lengths, we demonstrate that applying our reweighting framework reduces MSE by 10.1% on average and by up to 18.6% in the state-of-the-art model.
{"title":"Deep Imbalanced Time-series Forecasting via Local Discrepancy Density","authors":"Junwoo Park, Jungsoo Lee, Youngin Cho, W. Shin, Dongmin Kim, J. Choo, E. Choi","doi":"10.48550/arXiv.2302.13563","DOIUrl":"https://doi.org/10.48550/arXiv.2302.13563","url":null,"abstract":"Time-series forecasting models often encounter abrupt changes in a given period of time which generally occur due to unexpected or unknown events. Despite their scarce occurrences in the training set, abrupt changes incur loss that significantly contributes to the total loss. Therefore, they act as noisy training samples and prevent the model from learning generalizable patterns, namely the normal states. Based on our findings, we propose a reweighting framework that down-weights the losses incurred by abrupt changes and up-weights those by normal states. For the reweighting framework, we first define a measurement termed Local Discrepancy (LD) which measures the degree of abruptness of a change in a given period of time. Since a training set is mostly composed of normal states, we then consider how frequently the temporal changes appear in the training set based on LD. Our reweighting framework is applicable to existing time-series forecasting models regardless of the architectures. Through extensive experiments on 12 time-series forecasting models over eight datasets with various in-output sequence lengths, we demonstrate that applying our reweighting framework reduces MSE by 10.1% on average and by up to 18.6% in the state-of-the-art model.","PeriodicalId":74091,"journal":{"name":"Machine learning and knowledge discovery in databases : European Conference, ECML PKDD ... : proceedings. ECML PKDD (Conference)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"78848676","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-02-17DOI: 10.48550/arXiv.2302.09048
Clément Vignac, Nagham Osman, L. Toni, P. Frossard
This work introduces MiDi, a novel diffusion model for jointly generating molecular graphs and their corresponding 3D arrangement of atoms. Unlike existing methods that rely on predefined rules to determine molecular bonds based on the 3D conformation, MiDi offers an end-to-end differentiable approach that streamlines the molecule generation process. Our experimental results demonstrate the effectiveness of this approach. On the challenging GEOM-DRUGS dataset, MiDi generates 92% of stable molecules, against 6% for the previous EDM model that uses interatomic distances for bond prediction, and 40% using EDM followed by an algorithm that directly optimize bond orders for validity. Our code is available at github.com/cvignac/MiDi.
{"title":"MiDi: Mixed Graph and 3D Denoising Diffusion for Molecule Generation","authors":"Clément Vignac, Nagham Osman, L. Toni, P. Frossard","doi":"10.48550/arXiv.2302.09048","DOIUrl":"https://doi.org/10.48550/arXiv.2302.09048","url":null,"abstract":"This work introduces MiDi, a novel diffusion model for jointly generating molecular graphs and their corresponding 3D arrangement of atoms. Unlike existing methods that rely on predefined rules to determine molecular bonds based on the 3D conformation, MiDi offers an end-to-end differentiable approach that streamlines the molecule generation process. Our experimental results demonstrate the effectiveness of this approach. On the challenging GEOM-DRUGS dataset, MiDi generates 92% of stable molecules, against 6% for the previous EDM model that uses interatomic distances for bond prediction, and 40% using EDM followed by an algorithm that directly optimize bond orders for validity. Our code is available at github.com/cvignac/MiDi.","PeriodicalId":74091,"journal":{"name":"Machine learning and knowledge discovery in databases : European Conference, ECML PKDD ... : proceedings. ECML PKDD (Conference)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-02-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74653158","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：