-
公开(公告)号：US11775863B2
公开(公告)日：2023-10-03
申请号：US16781945
申请日：2020-02-04
Applicant: Oracle International Corporation
Abstract: Fairness of a trained classifier may be ensured by generating a data set for training, the data set generated using input data points of a feature space including multiple dimensions and according to different parameters including an amount of label bias, a control for discrepancy between rarity of features, and an amount of selection bias. Unlabeled data points of the input data comprising unobserved ground truths are labeled according to the amount of label bias and the input data sampled according to the amount of selection bias and the control for the discrepancy between the rarity of features. The classifier is then trained using the sampled and labeled data points as well as additional unlabeled data points. The trained classifier is then usable to determine unbiased classifications of one or more labels for one or more other data sets.
-
公开(公告)号：US20200372290A1
公开(公告)日：2020-11-26
申请号：US16781955
申请日：2020-02-04
Applicant: Oracle International Corporation
Inventor： Jean-Baptiste Frederic George Tristan , Pallika Haridas Kanani , Michael Louis Wick , Swetasudha Panda , Haniyeh Mahmoudian
Abstract: A Bayesian test of demographic parity for learning to rank may be applied to determine ranking modifications. A fairness control system receiving a ranking of items may apply Bayes factors to determine a likelihood of bias for the ranking. These Bayes factors may include a factor for determining bias in each item and a factor for determining bias in the ranking of the items. An indicator of bias may be generated using the applied Bayes factors and the fairness control system may modify the ranking if the determines likelihood of bias satisfies modification criteria for the ranking.
-
公开(公告)号：US20200226318A1
公开(公告)日：2020-07-16
申请号：US16833276
申请日：2020-03-27
Applicant: Oracle International Corporation
IPC: G06F40/137 , G06N7/00 , H03M7/30 , G06N5/04 , G06F17/16 , G06F40/146
Abstract: A scalable hierarchical coreference method that employs a homomorphic compression scheme that supports addition and partial subtraction to more efficiently represent the data and the evolving intermediate results of probabilistic inference. The method may encode the features underlying conditional random field models of coreference resolution so that cosine similarities can be efficiently computed. The method may be applied to compressing features and intermediate inference results for conditional random fields. The method may allow compressed representations to be added and subtracted in a way that preserves the cosine similarities.
-
公开(公告)号：US10606931B2
公开(公告)日：2020-03-31
申请号：US16379645
申请日：2019-04-09
Applicant: Oracle International Corporation
Abstract: A scalable hierarchical coreference method that employs a homomorphic compression scheme that supports addition and partial subtraction to more efficiently represent the data and the evolving intermediate results of probabilistic inference. The method may encode the features underlying conditional random field models of coreference resolution so that cosine similarities can be efficiently computed. The method may be applied to compressing features and intermediate inference results for conditional random fields. The method may allow compressed representations to be added and subtracted in a way that preserves the cosine similarities.
-
公开(公告)号：US20190354574A1
公开(公告)日：2019-11-21
申请号：US16379645
申请日：2019-04-09
Applicant: Oracle International Corporation
Abstract: A scalable hierarchical coreference method that employs a homomorphic compression scheme that supports addition and partial subtraction to more efficiently represent the data and the evolving intermediate results of probabilistic inference. The method may encode the features underlying conditional random field models of coreference resolution so that cosine similarities can be efficiently computed. The method may be applied to compressing features and intermediate inference results for conditional random fields. The method may allow compressed representations to be added and subtracted in a way that preserves the cosine similarities.
-
Patent Agency Ranking
公开(公告)号：US20240202612A1
公开(公告)日：2024-06-20
申请号：US18590285
申请日：2024-02-28
Applicant: Oracle International Corporation
IPC: G06Q10/04 , G02B27/01 , G06F16/2457 , G06F17/18 , G06F18/21 , G06F18/2113 , G06N20/00 , G06N20/20 , G06T19/00 , G06V20/20 , G09G3/00
CPC classification number: G06Q10/04 , G02B27/0101 , G02B27/0172 , G06F16/24578 , G06F17/18 , G06F18/2113 , G06F18/2193 , G06N20/00 , G06N20/20 , G06T19/006 , G06V20/20 , G09G3/003 , G02B2027/0118 , G02B2027/0138 , G02B2027/014 , G09G2320/0626
Abstract: A Bayesian test of demographic parity for learning to rank may be applied to determine ranking modifications. A fairness control system receiving a ranking of items may apply Bayes factors to determine a likelihood of bias for the ranking. These Bayes factors may include a factor for determining bias in each item and a factor for determining bias in the ranking of the items. An indicator of bias may be generated using the applied Bayes factors and the fairness control system may modify the ranking if the determines likelihood of bias satisfies modification criteria for the ranking.
-
公开(公告)号：US11948102B2
公开(公告)日：2024-04-02
申请号：US17819611
申请日：2022-08-12
Applicant: Oracle International Corporation
IPC: G06F16/2457 , G02B27/01 , G06F17/18 , G06F18/21 , G06F18/2113 , G06N20/00 , G06N20/20 , G06Q10/04 , G06T19/00 , G06V20/20 , G09G3/00
CPC classification number: G06Q10/04 , G02B27/0101 , G02B27/0172 , G06F16/24578 , G06F17/18 , G06F18/2113 , G06F18/2193 , G06N20/00 , G06N20/20 , G06T19/006 , G06V20/20 , G09G3/003 , G02B2027/0118 , G02B2027/0138 , G02B2027/014 , G09G2320/0626
Abstract: A Bayesian test of demographic parity for learning to rank may be applied to determine ranking modifications. A fairness control system receiving a ranking of items may apply Bayes factors to determine a likelihood of bias for the ranking. These Bayes factors may include a factor for determining bias in each item and a factor for determining bias in the ranking of the items. An indicator of bias may be generated using the applied Bayes factors and the fairness control system may modify the ranking if the determines likelihood of bias satisfies modification criteria for the ranking.
-
公开(公告)号：US20230401286A1
公开(公告)日：2023-12-14
申请号：US17903798
申请日：2022-09-06
Applicant: Oracle International Corporation
Inventor： Ariel Gedaliah Kobren , Swetasudha Panda , Michael Louis Wick , Qinlan Shen , Jason Anthony Peck
CPC classification number: G06K9/6257 , G06F40/56
Abstract: Techniques are disclosed for augmenting data sets used for training machine learning models and for generating predictions by trained machine learning models. These techniques may increase a number and diversity of examples within an initial training dataset of sentences by extracting a subset of words from the existing training dataset of sentences. The techniques may conserve scarce sample data in few-shot situations by training a data generation model using general data obtained from a general data source.
-
公开(公告)号：US20230368015A1
公开(公告)日：2023-11-16
申请号：US17940549
申请日：2022-09-08
Applicant: Oracle International Corporation
Inventor： Michael Louis Wick , Ariel Gedaliah Kobren , Swetasudha Panda
IPC: G06N3/08 , G06F40/279
CPC classification number: G06N3/08 , G06F40/279
Abstract: Techniques are described herein for training and applying machine learning models. The techniques include implementing an entropy-based loss function for training high-capacity machine learning models, such as deep neural networks, with anti-modeling. The entropy-based loss function may cause the model to have high entropy on negative data, helping prevent the model from becoming confidently wrong about the negative data while reducing the likelihood of generalizing from disfavored signals.
-
公开(公告)号：US20210375262A1
公开(公告)日：2021-12-02
申请号：US16890263
申请日：2020-06-02
Applicant: Oracle International Corporation
IPC: G10L15/01 , G10L15/197 , G10L15/06
Abstract: A method of evaluating a language model using negative data may include accessing a first language model that is trained using a first training corpus, and accessing a second language model. The second language model may be configured to generate outputs that are less grammatical than outputs generated by the first language model. The method may also include training the second language model using a second training corpus, and generating output text from the second language model. The method may further include testing the first language model using the output text from the second language model.
-
-
-
-
-
-
-
-
-
Search Results
34
records
(took 0.018 seconds)
