公开(公告)号：WO2023041145A1

公开(公告)日：2023-03-23

申请号：PCT/EP2021/075240

申请日：2021-09-14

Applicant: TELEFONAKTIEBOLAGET LM ERICSSON (PUBL)

Inventor： VUPPALA, Sunil, Kumar ,  SINGH, Saurabh ,  CYRAS, Kristijonas ,  MUJUMDAR, Anusha, Pradeep ,  SISODIA, Arpit ,  HU, Wenfeng ,  ANAMANDRA, Sai, Hareesh

IPC: G06N5/02 ,  G06N5/04 ,  G06N5/00 ,  G06N20/00

Abstract: There is provided a method for consolidating explanations associated with actions proposed based on a current state of a system and an intent. The method comprises acquiring (310) first and second explanations, the first and second explanations being associated with a proposed action or with different actions, wherein each of the first and second explanations includes one or more constraints, combining (320) constraints from the first and second explanations to form a set of constraints D, generating (330) a planning problem P = , wherein K consists of a set of predicates F and the set of constraints D, wherein A represents a set of possible actions, I represents an initial state of the system, G represents a goal state of the system, and Cost represents cost values associated with each constraint, and determining (340) a solution for the planning problem.

公开(公告)号：WO2021089429A2

公开(公告)日：2021-05-14

申请号：PCT/EP2020/080535

申请日：2020-10-30

Applicant: TELEFONAKTIEBOLAGET LM ERICSSON (PUBL)

Inventor： CHAWLA, Deepa ,  DIXIT, Gaurav ,  HU, Wenfeng ,  ICKIN, Selim ,  MORADI, Farnaz ,  SANDERS, Erik ,  SINGH, Saurabh ,  TAGHIA, Jalil ,  VANDIKAS, Konstantinos

IPC: G06N5/00 ,  G06N3/08 ,  G06N3/04

Abstract: A method of performing federated feature selection for a machine learning model in a federated learning environment includes obtaining, at a first resolution, a global set of selector neural network weights. At a second resolution, the method selects, for a plurality of first data subsets, a first set of features from a feature space by iteratively applying a first selector neural network that is initialized with the global set of selector neural network weights to the first data subset to obtain a first set of selector neural network weights. The first data subsets are divided into a plurality of second data subsets, and, at a third resolution, a second set of features is selected from the feature space.

公开(公告)号：WO2022182273A1

公开(公告)日：2022-09-01

申请号：PCT/SE2021/050164

申请日：2021-02-26

Applicant: TELEFONAKTIEBOLAGET LM ERICSSON (PUBL)

Inventor： TAGHIA, Jalil ,  HU, Wenfeng ,  VANDIKAS, Konstantinos ,  ICKIN, Selim

IPC: G06N20/00 ,  G06F9/54 ,  G06F12/02 ,  G06F15/167 ,  G06N3/04

Abstract: According to a second aspect, it is provided a method for enabling collaborative machine learning. The method is performed by an agent device. The method comprises the steps of: obtaining local input data; generating read interface parameters based on the local input data using a controller neural net being a first model; generating write interface parameters; transmitting a central reading request to the central device; receiving a central reading from the central device; updating the controller neural net of the agent device based on the central reading; and providing a predictor output of local input data based on the controller neural net and a second model of the agent device, the second model having as an input an output of the controller neural net, wherein the predictor output is obtained from the second model.

公开(公告)号：WO2020246918A1

公开(公告)日：2020-12-10

申请号：PCT/SE2019/050509

申请日：2019-06-03

Applicant: TELEFONAKTIEBOLAGET LM ERICSSON (PUBL)

Inventor： HU, Wenfeng ,  JEONG, Jaeseong ,  VERBULSKII, Vladimir ,  CORREIA, Rodrigo

IPC: H01Q3/00 ,  G06N3/02 ,  H04W16/28

Abstract: A network metrics repository stores cell performance metrics and rule-based data measured during operation of a communication network. A policy neural network circuit has an input layer having input nodes, a sequence of hidden layers, and at least one output node. A processor trains the policy neural network circuit to approximate a baseline rule-based policy for controlling a tilt angle of a remote electrical tilt (RET) antenna based on the rule-based data. The processor provides a live cell performance metric to input nodes, adapts weights that are used by the input nodes responsive to output of the output node, and controls operation of the tilt angle of the RET antenna based on the output The output node provides the output responsive to processing a stream of cell performance metrics through the input nodes. The processor controls operation of the RET antenna based on the output.

公开(公告)号：WO2023088593A1

公开(公告)日：2023-05-25

申请号：PCT/EP2022/076035

申请日：2022-09-20

Applicant: TELEFONAKTIEBOLAGET LM ERICSSON (PUBL)

Inventor： HU, Wenfeng ,  VANDIKAS, Konstantinos ,  MENDO MATEO, Adriano ,  ISAKSSON, Martin ,  SANDERS, Erik

IPC: H04W24/02 ,  G06N3/045

Abstract: Embodiments herein relate, in some examples, to a method performed by a first radio network node (11) for handling data of a RAN in a communication network. The first radio network node obtains, from a second radio network node (12), a matrix indication of a second local computation associated with a GNN for predicting characteristics of the RAN, wherein the matrix indication is obtained over an internal interface when the first and second network node are comprised in a same logical radio network node, or the matrix indication is obtained over an external interface when the first and second network node are separated neighbouring radio network nodes. The first radio network node executes a first local computation, associated with the GNN for predicting the characteristics of the RAN, based on the obtained matrix indication, wherein an output of the first local computation indicates a gradient; and sends an indication of the gradient to a central network node (13) training the GNN for predicting the characteristics of the RAN.

公开(公告)号：WO2022248040A1

公开(公告)日：2022-12-01

申请号：PCT/EP2021/064101

申请日：2021-05-26

Applicant: TELEFONAKTIEBOLAGET LM ERICSSON (PUBL)

Inventor： JEONG, Jaeseong ,  DEMIREL, Burak ,  TATED, Harshit ,  HU, Wenfeng

IPC: G06N3/04 ,  G06N3/08 ,  H04W72/04

Abstract: The present disclosure provides a computer-implemented method for determining whether to perform an action proposed by a model. The model is developed using a reinforcement learning process. The method comprises classifying at least one of a plurality of inputs to the model as being supportive or resistant to an action proposed by the model. The method further comprises comparing the classification of the at least one of the plurality of inputs to domain knowledge to determine whether or not the proposed action conflicts with the domain knowledge, and, in response to determining that the proposed action does not conflict with the domain knowledge, initiating the proposed action. In this context, the domain knowledge is indicative of a relationship between the proposed action and the at least one of the plurality of inputs.

公开(公告)号：WO2022180421A1

公开(公告)日：2022-09-01

申请号：PCT/IB2021/051576

申请日：2021-02-25

Applicant: TELEFONAKTIEBOLAGET LM ERICSSON (PUBL)

Inventor： TAGHIA, Jalil ,  HU, Wenfeng ,  LEE, Carmen

IPC: G06N3/04 ,  G06N3/08

Abstract: A computer implemented method (100) is disclosed for generating a Machine Learning (ML) composition that is optimized to perform a composition task in a communication network. The ML composition comprises a plurality of interconnected ML modules, each ML module trained to perform a module task that is specific to the ML module, and an ML module comprises at least one ML model. The method comprises obtaining a candidate set of ML modules (110) and initiating a current version of a topology for the composition (120). The method further comprises repeating, until a termination condition is satisfied, the steps of identifying, from the candidate set of ML modules, a possible topology for the composition that includes the current version of the topology and minimizes a first loss function (130), evolving the current version of the topology based on the identified possible topology (140), and evaluating the current version of the topology using a second loss function (150). The ML composition optimized to perform the composition task comprises the ML modules and connections between ML modules present in the current version of the topology when the termination condition is satisfied.

公开(公告)号：WO2018206504A1

公开(公告)日：2018-11-15

申请号：PCT/EP2018/061716

申请日：2018-05-07

Applicant: TELEFONAKTIEBOLAGET LM ERICSSON (PUBL)

Inventor： HUANG, Vincent ,  HU, Wenfeng ,  LEY, Tobias ,  VLACHOU-KONCHYLAKI, Martha

IPC: G06N3/04 ,  G06N3/08

Abstract: A pre-training apparatus and method for reinforcement learning based on a Generative Adversarial Network (GAN) is provided. GAN includes a generator and a discriminator. The method comprising receiving training data from a real environment where the training data includes a data slice corresponding to a first state-reward pair and a first state-action pair, training the GAN using the training data, training a relations network to extract a latent relationship of the first state-action pair with the first state-reward pair in a reinforcement learning context, causing the generator trained with training data to generate first synthetic data, processing a portion of the first synthetic data in the relations network to generate a resulting data slice, merging the second state-action pair portion of the first synthetic data with the second state-reward pair from the relations network to generate second synthetic data to update a policy for interaction with the real environment.

公开(公告)号：WO2022093084A1

公开(公告)日：2022-05-05

申请号：PCT/SE2020/051041

申请日：2020-10-28

Applicant: TELEFONAKTIEBOLAGET LM ERICSSON (PUBL)

Inventor： WANG, Yu ,  HU, Wenfeng ,  SAXENA, Vidit ,  SOLDATI, Pablo

IPC: H04W24/00

Abstract: A method performed by a central node for controlling an exploration strategy associated to Reinforcement Learning, RL, in one or more RL modules in a distributed node in a Radio Access Network, RAN, is provided. The central node evaluates (401) a cost of actions performed for explorations in the one or more RL modules, and a performance of the one or more RL modules. Based on the evaluation, the central node determines (402) one or more exploration parameters associated to the exploration strategy. The central node controls the exploration strategy by configuring (403) the one or more RL modules with the determined one or more exploration parameters to update its exploration strategy, enforcing the respective one or more RL modules to act according to the updated exploration strategy to produce data samples for the one or more RL modules in the distributed node.

公开(公告)号：WO2021249648A1

公开(公告)日：2021-12-16

申请号：PCT/EP2020/066235

申请日：2020-06-11

Applicant: TELEFONAKTIEBOLAGET LM ERICSSON (PUBL)

Inventor： MORADI, Farnaz ,  VANDIKAS, Konstantinos ,  ICKIN, Selim ,  HU, Wenfeng ,  TAGHIA, Jalil

IPC: G06N3/04 ,  G06N3/08 ,  G06N20/20 ,  G06N3/0454

Abstract: Methods, systems, and apparatuses are presented for grouping worker nodes in a machine learning system comprising a master node and a plurality of worker nodes, the method comprising grouping each worker node of the plurality of worker nodes into a group of a plurality of groups based on characteristics of a data distribution of each of the plurality of worker nodes, subgrouping worker nodes within the group of the plurality of groups into subgroups based on characteristics of a worker neural network model of each worker node from the group of the plurality of groups, averaging the worker neural network models of worker nodes within a subgroup to generate a subgroup average model, and distributing the subgroup average model.