研究兴趣

论文共 20 篇作者统计合作学者相似作者

按年份排序按引用量排序主题筛选期刊级别筛选合作者筛选合作机构筛选
时间
引用量
主题
期刊级别
合作者
合作机构
Knowledge Propagation over Conditional Independence Graphs
Urszula Chajewska,Harsh Shrivastava
CoRR (2023)
引用0浏览0EI引用
0
0
Are uGLAD? Time will tell
Shima Imani,Harsh Shrivastava
arxiv（2023）
引用0浏览0EI引用
0
0
Neural Graph Revealers
Harsh Shrivastava, Urszula Chajewska
arxiv（2023）
引用0浏览0引用
0
0
uGLAD: Sparse graph recovery by optimizing deep unrolled networks
Harsh Shrivastava,Urszula Chajewska, Robin Abraham, Xinshi Chen
arxiv（2022）
引用0浏览0引用
0
0
Neural Graphical Models
Harsh Shrivastava, Urszula Chajewska
ICLR 2023 (2022): 284-307
引用0浏览0EI引用
0
0
加载更多
作者统计
  • 合作者
  • 学生
  • 导师
暂无相似学者，你可以通过学者研究领域进行搜索筛选
数据免责声明
页面数据均来自互联网公开来源、合作出版商和通过AI技术自动分析结果，我们不对页面数据的有效性、准确性、正确性、可靠性、完整性和及时性做出任何承诺和保证。若有疑问，可以通过电子邮件方式联系我们：report@aminer.cn
which draws a parallel between the CI graph nodes and the variables of the time series. Consider applying a graph recovery model $\\texttt{uGLAD}$ to a short interval of the time series, it will result in a CI graph that shows partial correlations among the variables. We extend this idea to the entire time series by utilizing a sliding window to create a batch of time intervals and then run a single $\\texttt{uGLAD}$ model in multitask learning mode to recover all the CI graphs simultaneously. As a result, we obtain a corresponding temporal CI graphs representation. We then designed a first-order and second-order based trajectory tracking algorithms to study the evolution of these graphs across distinct intervals. Finally, an `Allocation' algorithm is used to determine a suitable segmentation of the temporal graph sequence. $\\texttt{tGLAD}$ provides a competitive time complexity of $O(N)$ for settings where number of variables $D\u003C\u003CN$. We demonstrate successful empirical results on a Physical Activity Monitoring data. ","authors":[{"name":"Shima Imani"},{"id":"63808126fc451b2d602afcf5","name":"Harsh Shrivastava"}],"create_time":"2023-03-22T05:02:31.005Z","hashs":{"h1":"- uglad time"},"id":"641a71fb90e50fcafd71ffcf","lang":"en","num_citation":0,"pdf":"https:\u002F\u002Fcz5waila03cyo0tux1owpyofgoryroob.aminer.cn\u002F4D\u002F35\u002F60\u002F4D3560815C1A1995E863DE71C5BBEF3D.pdf","pdf_src":["https:\u002F\u002Farxiv.org\u002Fpdf\u002F2303.11647"],"title":"Are uGLAD? Time will tell!","update_times":{"u_a_t":"2023-03-24T12:32:18.32Z"},"urls":["db\u002Fjournals\u002Fcorr\u002Fcorr2303.html#abs-2303-11647","https:\u002F\u002Fdoi.org\u002F10.48550\u002FarXiv.2303.11647","https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.11647"],"versions":[{"id":"641a71fb90e50fcafd71ffcf","sid":"2303.11647","src":"arxiv","year":2023},{"id":"6427029d90e50fcafd5d928e","sid":"journals\u002Fcorr\u002Fabs-2303-11647","src":"dblp","year":2023}],"year":2023},{"abstract":" Sparse graph recovery methods work well where the data follows their assumptions but often they are not designed for doing downstream probabilistic queries. This limits their adoption to only identifying connections among the input variables. On the other hand, the Probabilistic Graphical Models (PGMs) assume an underlying base graph between variables and learns a distribution over them. PGM design choices are carefully made such that the inference \\& sampling algorithms are efficient. This brings in certain restrictions and often simplifying assumptions. In this work, we propose Neural Graph Revealers (NGRs), that are an attempt to efficiently merge the sparse graph recovery methods with PGMs into a single flow. The problem setting consists of an input data X with D features and M samples and the task is to recover a sparse graph showing connection between the features and learn a probability distribution over the D at the same time. NGRs view the neural networks as a `glass box' or more specifically as a multitask learning framework. We introduce `Graph-constrained path norm' that NGRs leverage to learn a graphical model that captures complex non-linear functional dependencies between the features in the form of an undirected sparse graph. Furthermore, NGRs can handle multimodal inputs like images, text, categorical data, embeddings etc. which is not straightforward to incorporate in the existing methods. We show experimental results of doing sparse graph recovery and probabilistic inference on data from Gaussian graphical models and a multimodal infant mortality dataset by Centers for Disease Control and Prevention. ","authors":[{"id":"63808126fc451b2d602afcf5","name":"Harsh Shrivastava"},{"name":"Urszula Chajewska"}],"create_time":"2023-02-28T04:55:05.989Z","hashs":{"h1":"-neural graph revealers"},"id":"63fd715990e50fcafd146f65","lang":"en","num_citation":0,"pdf":"https:\u002F\u002Fcz5waila03cyo0tux1owpyofgoryroob.aminer.cn\u002FB7\u002FA5\u002F28\u002FB7A528A5F97CAA90BE8BEEC9DA2CAEDE.pdf","pdf_src":["https:\u002F\u002Farxiv.org\u002Fpdf\u002F2302.13582"],"title":"Neural Graph Revealers","urls":["https:\u002F\u002Farxiv.org\u002Fabs\u002F2302.13582"],"versions":[{"id":"63fd715990e50fcafd146f65","sid":"2302.13582","src":"arxiv","year":2023}],"year":2023},{"abstract":" Large Language Models (LLMs) have limited performance when solving arithmetic reasoning tasks and often provide incorrect answers. Unlike natural language understanding, math problems typically have a single correct answer, making the task of generating accurate solutions more challenging for LLMs. To the best of our knowledge, we are not aware of any LLMs that indicate their level of confidence in their responses which fuels a trust deficit in these models impeding their adoption. To address this deficiency, we propose `MathPrompter', a technique that improves performance of LLMs on arithmetic problems along with increased reliance in the predictions. MathPrompter uses the Zero-shot chain-of-thought prompting technique to generate multiple Algebraic expressions or Python functions to solve the same math problem in different ways and thereby raise the confidence level in the output results. This is in contrast to other prompt based CoT methods, where there is no check on the validity of the intermediate steps followed. Our technique improves over state-of-the-art on the MultiArith dataset ($78.7\\%\\rightarrow92.5\\%$) evaluated using 175B parameter GPT-based LLM. ","authors":[{"id":"53f42fc4dabfaedf435349de","name":"Shima Imani"},{"name":"Liang Du"},{"id":"63808126fc451b2d602afcf5","name":"Harsh Shrivastava"}],"create_time":"2023-03-10T04:54:43.431Z","flags":[{"flag":"affirm_author","person_id":"637381699bb5705eda8ac8b9"}],"hashs":{"h1":"mmrll","h3":"m"},"id":"640a9ffc90e50fcafd03ca54","lang":"en","num_citation":49,"pages":{"end":"42","start":"37"},"pdf":"https:\u002F\u002Fstatic.aminer.cn\u002Fupload\u002Fpdf\u002F1745\u002F1132\u002F1722\u002F640a9ffc90e50fcafd03ca54_0.pdf","pdf_src":["https:\u002F\u002Farxiv.org\u002Fpdf\u002F2303.05398"],"title":"MathPrompter: Mathematical Reasoning using Large Language Models","update_times":{"u_a_t":"2023-03-10T05:18:55.531Z","u_c_t":"2023-11-26T16:31:43.989Z","u_v_t":"2023-11-18T14:55:03.312Z"},"urls":["db\u002Fconf\u002Facl\u002Facl2023i.html#ImaniD023","https:\u002F\u002Faclanthology.org\u002F2023.acl-industry.4","https:\u002F\u002Faclanthology.org\u002F2023.acl-industry.4\u002F","https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.05398","db\u002Fjournals\u002Fcorr\u002Fcorr2303.html#abs-2303-05398","https:\u002F\u002Fdoi.org\u002F10.48550\u002FarXiv.2303.05398"],"venue":{"info":{"name":"conf_acl"},"volume":"abs\u002F2303.05398"},"venue_hhb_id":"5ea1afddedb6e7d53c00c104","versions":[{"id":"640a9ffc90e50fcafd03ca54","sid":"2303.05398","src":"arxiv","year":2023},{"id":"6427029d90e50fcafd5da172","sid":"journals\u002Fcorr\u002Fabs-2303-05398","src":"dblp","year":2023},{"id":"64ae66c23fda6d7f068474a5","sid":"2023.acl-industry.4","src":"conf_acl","year":2023},{"id":"64c78b9f3fda6d7f06db990c","sid":"conf\u002Facl\u002FImaniD023","src":"dblp","year":2023}],"year":2023},{"abstract":"Probabilistic Graphical Models are generative models of complex systems. They rely on conditional independence assumptions between variables to learn sparse representations which can be visualized in a form of a graph. Such models are used for domain exploration and structure discovery in poorly understood domains. This work introduces a novel technique to perform sparse graph recovery by optimizing deep unrolled networks. Assuming that the input data $X\\in\\mathbb{R}^{M\\times D}$ comes from an underlying multivariate Gaussian distribution, we apply a deep model on $X$ that outputs the precision matrix $\\Theta$. Then, the partial correlation matrix \\mathrm{P} is calculated which can also be interpreted as providing a list of conditional independence assertions holding in the input distribution. Our model, \\texttt{uGLAD}, builds upon and extends the state-of-the-art model \\texttt{GLAD} to the unsupervised setting. The key benefits of our model are (1) \\texttt{uGLAD} automatically optimizes sparsity-related regularization parameters leading to better performance than existing algorithms. (2) We introduce multi-task learning based `consensus' strategy for robust handling of missing data in an unsupervised setting. We evaluate performance on synthetic Gaussian, non-Gaussian data generated from Gene Regulatory Networks, and present case studies in anaerobic digestion and infant mortality.","authors":[{"id":"63808126fc451b2d602afcf5","name":"Harsh Shrivastava","org":"Microsoft","orgid":"5f71b2831c455f439fe3c634","orgs":["Microsoft"]},{"name":"Urszula Chajewska","org":"Microsoft","orgid":"5f71b2831c455f439fe3c634","orgs":["Microsoft"]},{"id":"53f4ccacdabfaeebd9f8163f","name":"Robin Abraham","org":"Microsoft","orgid":"5f71b2831c455f439fe3c634","orgs":["Microsoft"]},{"id":"561d1ba845ce1e596473fabf","name":"Xinshi Chen","org":"Georgia Institute of Technology","orgid":"5f71b4f61c455f439fe4dc1a","orgs":["Georgia Institute of Technology"]}],"create_time":"2023-04-11T03:08:55.505Z","hashs":{"h1":"udlmr","h3":"cig"},"id":"6433f69090e50fcafd6e124f","keywords":["Graphical Lasso","Deep Learning","Unrolled Algorithms","Conditional Independence graphs","Sparse graphs"],"lang":"en","num_citation":0,"pdf_src":["https:\u002F\u002Fopenreview.net\u002Fpdf\u002Fbc80a9227fea49be8184c7c7b8337a59e740d921.pdf"],"title":"uGLAD: A deep learning model to recover conditional independence graphs","urls":["https:\u002F\u002Fopenreview.net\u002Fforum?id=dmWMfJeZMM"],"venue":{"info":{"name":"ICLR 2023"}},"versions":[{"id":"6433f69090e50fcafd6e124f","sid":"2023#dmWMfJeZMM","src":"conf_iclr","vsid":"ICLR.cc\u002F2023\u002FConference","year":2023}],"year":2023},{"abstract":" In this paper, we introduce DiversiGATE, a unified framework that consolidates diverse methodologies for LLM verification. The proposed framework comprises two main components: Diversification and Aggregation which provide a holistic perspective on existing verification approaches, such as Self-Consistency, Math Prompter and WebGPT. Furthermore, we propose a novel `SelfLearner' model that conforms to the DiversiGATE framework which can learn from its own outputs and refine its performance over time, leading to improved accuracy. To evaluate the effectiveness of SelfLearner, we conducted a rigorous series of experiments, including tests on synthetic data as well as on popular arithmetic reasoning benchmarks such as GSM8K. Our results demonstrate that our approach outperforms traditional LLMs, achieving a considerable 54.8% -\u003E 61.8% improvement on the GSM8K benchmark. ","authors":[{"id":"53f42fc4dabfaedf435349de","name":"Shima Imani"},{"name":"Ali Beyram"},{"id":"63808126fc451b2d602afcf5","name":"Harsh Shrivastava"}],"create_time":"2023-06-26T04:49:22.461Z","hashs":{"h1":"dcfrl","h3":"lm"},"id":"64990ccbd68f896efaf84759","num_citation":1,"pdf":"https:\u002F\u002Fcz5waila03cyo0tux1owpyofgoryroob.aminer.cn\u002FB2\u002FC1\u002F99\u002FB2C199AC557A0583C74EDC4736C9E55B.pdf","title":"DiversiGATE: A Comprehensive Framework for Reliable Large Language\n Models","update_times":{"u_c_t":"2023-11-03T07:38:37.204Z"},"urls":["db\u002Fjournals\u002Fcorr\u002Fcorr2306.html#abs-2306-13230","https:\u002F\u002Fdoi.org\u002F10.48550\u002FarXiv.2306.13230","https:\u002F\u002Farxiv.org\u002Fabs\u002F2306.13230"],"venue":{"info":{"name":"CoRR"},"volume":"abs\u002F2306.13230"},"versions":[{"id":"64990ccbd68f896efaf84759","sid":"2306.13230","src":"arxiv","year":2023},{"id":"64a29646d68f896efa296ca8","sid":"journals\u002Fcorr\u002Fabs-2306-13230","src":"dblp","year":2023}],"year":2023},{"abstract":" Probabilistic Graphical Models (PGMs) are generative models of complex systems. They rely on conditional independence assumptions between variables to learn sparse representations which can be visualized in a form of a graph. Such models are used for domain exploration and structure discovery in poorly understood domains. This work introduces a novel technique to perform sparse graph recovery by optimizing deep unrolled networks. Assuming that the input data $X\\in\\mathbb{R}^{M\\times D}$ comes from an underlying multivariate Gaussian distribution, we apply a deep model on $X$ that outputs the precision matrix $\\hat{\\Theta}$, which can also be interpreted as the adjacency matrix. Our model, uGLAD, builds upon and extends the state-of-the-art model GLAD to the unsupervised setting. The key benefits of our model are (1) uGLAD automatically optimizes sparsity-related regularization parameters leading to better performance than existing algorithms. (2) We introduce multi-task learning based `consensus' strategy for robust handling of missing data in an unsupervised setting. We evaluate model results on synthetic Gaussian data, non-Gaussian data generated from Gene Regulatory Networks, and present a case study in anaerobic digestion. ","authors":[{"id":"63808126fc451b2d602afcf5","name":"Harsh Shrivastava"},{"id":"5d415bed7390bff0db70aeb0","name":"Urszula Chajewska"},{"name":"Robin Abraham"},{"name":"Xinshi Chen"}],"create_time":"2022-05-25T13:44:42.663Z","hashs":{"h1":"usgro","h3":"dun"},"id":"628d9e795aee126c0f9791a2","lang":"en","num_citation":0,"pdf":"https:\u002F\u002Fcz5waila03cyo0tux1owpyofgoryroob.aminer.cn\u002F1A\u002F82\u002F27\u002F1A8227F7CDBE9292161C2A0410C1446E.pdf","pdf_src":["https:\u002F\u002Farxiv.org\u002Fpdf\u002F2205.11610"],"title":"uGLAD: Sparse graph recovery by optimizing deep unrolled networks","urls":["https:\u002F\u002Farxiv.org\u002Fabs\u002F2205.11610"],"versions":[{"id":"628d9e795aee126c0f9791a2","sid":"2205.11610","src":"arxiv","year":2022}],"year":2022},{"abstract":" Graphs are ubiquitous and are often used to understand the dynamics of a system. Probabilistic Graphical Models comprising Bayesian and Markov networks, and Conditional Independence graphs are some of the popular graph representation techniques. They can model relationships between features (nodes) together with the underlying distribution. Although theoretically these models can represent very complex dependency functions, in practice often simplifying assumptions are made due to computational limitations associated with graph operations. This work introduces Neural Graphical Models (NGMs) which attempt to represent complex feature dependencies with reasonable computational costs. Specifically, given a graph of feature relationships and corresponding samples, we capture the dependency structure between the features along with their complex function representations by using neural networks as a multi-task learning framework. We provide efficient learning, inference and sampling algorithms for NGMs. Moreover, NGMs can fit generic graph structures including directed, undirected and mixed-edge graphs as well as support mixed input data types. We present empirical studies that show NGMs' capability to represent Gaussian graphical models, inference analysis of a lung cancer data and extract insights from a real world infant mortality data provided by CDC. ","authors":[{"id":"63808126fc451b2d602afcf5","name":"Harsh Shrivastava","org":"Microsoft","orgid":"5f71b2831c455f439fe3c634","orgs":["Microsoft"]},{"name":"Urszula Chajewska","org":"Microsoft","orgid":"5f71b2831c455f439fe3c634","orgs":["Microsoft"]}],"create_time":"2022-10-13T05:01:28.833Z","hashs":{"h1":"-neural graphical models"},"id":"633ba44890e50fcafdfe4dfd","keywords":["Graphical models","Deep learning","Learning Representations"],"lang":"en","num_citation":0,"pages":{"end":"307","start":"284"},"pdf":"https:\u002F\u002Fcz5waila03cyo0tux1owpyofgoryroob.aminer.cn\u002F74\u002F15\u002FFB\u002F7415FB12241E63ABB77EC0A1F1FC2B33.pdf","pdf_src":["https:\u002F\u002Farxiv.org\u002Fpdf\u002F2210.00453","https:\u002F\u002Fopenreview.net\u002Fpdf?id=UA34f_shAO","https:\u002F\u002Fopenreview.net\u002Fpdf\u002F34d08dbaae8304bf5dd79e73f6484f80095d8aa1.pdf"],"title":"Neural Graphical Models","update_times":{"u_v_t":"2023-04-11T03:08:55.505Z"},"urls":["https:\u002F\u002Fdl.acm.org\u002Fdoi\u002F10.1007\u002F978-3-031-45608-4_22","db\u002Fjournals\u002Fcorr\u002Fcorr2309.html#abs-2309-11680","https:\u002F\u002Fdoi.org\u002F10.48550\u002FarXiv.2309.11680","https:\u002F\u002Farxiv.org\u002Fabs\u002F2309.11680","https:\u002F\u002Farxiv.org\u002Fabs\u002F2210.00453","https:\u002F\u002Fopenreview.net\u002Fforum?id=UA34f_shAO"],"venue":{"info":{"name":"ICLR 2023"},"volume":"abs\u002F2309.11680"},"versions":[{"id":"633ba44890e50fcafdfe4dfd","sid":"2210.00453","src":"arxiv","year":2022},{"id":"63dcdb422c26941cf00b6b23","sid":"UA34f_shAO","src":"conf_iclr","vsid":"ICLR.cc\u002F2023\u002FConference","year":2023},{"id":"6433f68f90e50fcafd6e10aa","sid":"2023#UA34f_shAO","src":"conf_iclr","vsid":"ICLR.cc\u002F2023\u002FConference","year":2023},{"id":"650cf9223fda6d7f06d42a86","sid":"2309.11680","src":"arxiv","year":2023},{"id":"6523b754939a5f40821c1d99","sid":"journals\u002Fcorr\u002Fabs-2309-11680","src":"dblp","year":2023},{"id":"656404a2939a5f408223267e","sid":"10.1007\u002F978-3-031-45608-4_22","src":"acm","year":2023}],"year":2022},{"abstract":"Motivation: Reconstruction of genome-scale networks from gene expression data is an actively studied problem. A wide range of methods that differ between the types of interactions they uncover with varying trade-offs between sensitivity and specificity have been proposed. To leverage benefits of multiple such methods, ensemble network methods that combine predictions from resulting networks have been developed, promising results better than or as good as the individual networks. Perhaps owing to the difficulty in obtaining accurate training examples, these ensemble methods hitherto are unsupervised. Results: In this article, we introduce EnGRaiN, the first supervised ensemble learning method to construct gene networks. The supervision for training is provided by small training datasets of true edge connections (positives) and edges known to be absent (negatives) among gene pairs. We demonstrate the effectiveness of EnGRaiN using simulated datasets as well as a curated collection of Arabidopsis thaliana datasets we created from microarray datasets available from public repositories. EnGRaiN shows better results not only in terms of receiver operating characteristic and PR characteristics for both real and simulated datasets compared with unsupervised methods for ensemble network construction, but also generates networks that can be mined for elucidating complex biological interactions.","authors":[{"id":"5626240b45ce1e5965337e84","name":"Maneesha Aluru","org":"Georgia Inst Technol, Dept Biol, Atlanta, GA 30308 USA","orgs":["Georgia Inst Technol, Dept Biol, Atlanta, GA 30308 USA"]},{"id":"63808126fc451b2d602afcf5","name":"Harsh Shrivastava","org":"Microsoft, Redmond, WA 98052 USA","orgid":"5f71b2831c455f439fe3c634","orgs":["Microsoft, Redmond, WA 98052 USA"]},{"id":"53f439b1dabfaee4dc7a13d4","name":"Sriram P Chockalingam","org":"Georgia Inst Technol, Inst Data Engn & Sci, Atlanta, GA 30308 USA","orgs":["Georgia Inst Technol, Inst Data Engn & Sci, Atlanta, GA 30308 USA"]},{"id":"64972993a88fbe7c05ffe084","name":"Shruti Shivakumar","org":"Georgia Inst Technol, Dept Computat Sci & Engn, Atlanta, GA 30308 USA","orgs":["Georgia Inst Technol, Dept Computat Sci & Engn, Atlanta, GA 30308 USA"]},{"id":"53f4314fdabfaedce54f78e0","name":"Srinivas Aluru","org":"Georgia Inst Technol, Inst Data Engn & Sci, Atlanta, GA 30308 USA","orgs":["Georgia Inst Technol, Inst Data Engn & Sci, Atlanta, GA 30308 USA","Georgia Inst Technol, Dept Computat Sci & Engn, Atlanta, GA 30308 USA"]}],"citations":{"google_citation":4,"last_citation":4},"create_time":"2021-12-11T14:49:45.712Z","doi":"10.1093\u002Fbioinformatics\u002Fbtab829","flags":[{"flag":"affirm_author","person_id":"53f4314fdabfaedce54f78e0"}],"hashs":{"h1":"eselm","h3":"rlgrn"},"id":"61b480c15244ab9dcbe9c5ac","issn":"1367-4803","lang":"en","num_citation":8,"pages":{"end":"1319","start":"1312"},"title":"EnGRaiN: a supervised ensemble learning method for recovery of large-scale gene regulatory networks","update_times":{"u_c_t":"2023-03-29T10:30:50.083Z","u_v_t":"2022-05-26T02:23:56.794Z"},"urls":["https:\u002F\u002Fdoi.org\u002F10.1093\u002Fbioinformatics\u002Fbtab829","https:\u002F\u002Fwww.ncbi.nlm.nih.gov\u002Fpubmed\u002F34888624","http:\u002F\u002Fwww.webofknowledge.com\u002F","db\u002Fjournals\u002Fbioinformatics\u002Fbioinformatics38.html#AluruSCSA22"],"venue":{"info":{"name":"BIOINFORMATICS"},"issue":"5","volume":"38"},"venue_hhb_id":"5ea182aaedb6e7d53c009751","versions":[{"id":"61b480c15244ab9dcbe9c5ac","sid":"34888624","src":"pubmed","vsid":"9808944","year":2022},{"id":"623b118d5aee126c0fdb6d11","sid":"journals\u002Fbioinformatics\u002FAluruSCSA22","src":"dblp","vsid":"journals\u002Fbioinformatics","year":2022},{"id":"628d217e5aee126c0f432915","sid":"WOS:000776280200017","src":"wos","vsid":"BIOINFORMATICS","year":2022}],"year":2022}],"profilePubsTotal":20,"profilePatentsPage":0,"profilePatents":null,"profilePatentsTotal":null,"profilePatentsEnd":false,"profileProjectsPage":1,"profileProjects":{"success":true,"msg":"","data":null,"log_id":"2Yy9cMY6TlVJyUAsTJISYnL3vt1"},"profileProjectsTotal":0,"newInfo":null,"checkDelPubs":[]}};