Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for selecting an input vocabulary for a machine learning model using power indices. One of the methods includes computing a respective score for each of a plurality of text tokens in an initial vocabulary and then selecting the text tokens in the input vocabulary based on the respective scores.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for selecting an optimal feature point in a continuous domain for a group of agents. A computer-implemented system obtains, for each of a plurality of agents, respective training data that comprises a respective utility score for each of a plurality of discrete points in the continuous domain. The system trains, for each of the plurality of agents and on the respective training data for the agents, a respective neural network that is configured to receive an input comprising a point in the continuous domain and to generate as output a predicted utility score for the agent at the point.

Abstract: Methods, systems and apparatus, including computer programs encoded on computer storage media, for training a policy neural network by repeatedly updating the policy neural network at each of a plurality of training iterations. One of the methods includes generating training data for the training iteration by controlling the agent in accordance with an improved policy that selects actions in response to input state representations. A best response computation is performed using (i) a candidate policy generated from respective policy neural networks as of one or more preceding iterations and (ii) a candidate value neural network. The candidate value neural network is configured to generate a value output that is an estimate of a value of the environment being in the state characterized by a state representation to complete a particular task. The policy neural network is updated by training the policy neural network on the training data.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for efficiently allocating resources among participants. Methods can include receiving valuation data specifying, for each of a plurality of entities, a respective valuation for each of a plurality of resource subsets, each resource subset comprising a different combination of one or more resources of a plurality of resources. After receiving valuation data, assigning each resource in the plurality of resources to a respective entity of the plurality of entities based on the valuations and generating, for each particular entity, a respective input representation that is derived from valuations of every other entity in the plurality of entities other than the particular entity. The input representation for each particular entity is processed using a neural network to generate a rule for the particular entity and a payment based on the rule output for the entities.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for efficiently allocating resources among participants. Methods can include receiving valuation data specifying, for each of a plurality of entities, a respective valuation for each of a plurality of resource subsets, each resource subset comprising a different combination of one or more resources of a plurality of resources. After receiving valuation data, assigning each resource in the plurality of resources to a respective entity of the plurality of entities based on the valuations and generating, for each particular entity, a respective input representation that is derived from valuations of every other entity in the plurality of entities other than the particular entity. The input representation for each particular entity is processed using a neural network to generate a rule for the particular entity and a payment based on the rule output for the entities.

Abstract: Checking and/or completing for data grids is described such as for grids having rows and columns of cells at least some of which contain data values such as numbers or categories. In various embodiments predictive probability distributions are obtained from an inference engine for one or more of the cells and the predictive probability distributions are used for various tasks such as to suggest values to complete blank cells, highlight cells having outlying values, identify potential errors, suggest corrections to potential errors, identify similarities between cells, identify differences between cells, cluster rows of the data grid, and other tasks. In various embodiments a graphical user interface displays a data grid and provides facilities for completing, error checking/correcting, and analyzing data in the data grid.

Abstract: A computer-implemented method of tagging a text, comprises: determining a value for each of a plurality of locations in a first vector; processing (402), by a trained first neural network component, the first vector to generate a second vector; processing (404), at a trained second neural network component, the second vector to generate a probability score for each of at least ten predetermined tags; determining (406) if each probability score meets a criterion; if the criterion is met, assigning (408) the tag corresponding to the probability score to the text. Each of the locations may correspond to a respective predetermined word, each value relating to existence and/or frequency of the corresponding word in the text, and the number of locations may be between 600 and 20000. The number of locations in the second vector may be fewer than the number of locations in the first vector and is from 100 to 5000.

Abstract: Certain examples described herein provide methods and systems for implementing a conversational agent, e.g. to train a predictive model used by the conversational agent. In examples, text data representing agent messages from a dialogue database are clustered and the clusters are used to generate response templates for use by the conversational agent. The predictive model is trained on training data generated by selectively assigning response templates to agent messages from text dialogues. Examples enable a predictive model to be trained on high quality data sets that are generated automatically from a corpus of historical data. In turn, they enable a natural language interface to be efficiently provided.

Abstract: Certain examples are described that provide methods and systems for generating templates for use by a conversational agent. These examples enable a natural language interface to be provided. Certain examples cluster user and agent messages from a corpus of text data representing text dialogues. This clustering enables response templates to be generated in a way that takes into account a context in which responses are given. In certain examples, messages that are exchanged between a user and a conversational agent are embedded as numeric arrays based a neural sequence-to-sequence model. Clustering routines are used to group dialogue encodings into one or more response clusters, and these clusters may then be used to generate response templates. The response templates may be used by a conversational agent to prepare a response to a user message.

Abstract: Certain examples described herein provide methods and systems for implementing a conversational agent, e.g. to train a predictive model used by the conversational agent. In examples, text data representing agent messages from a dialogue database are clustered and the clusters are used to generate response templates for use by the conversational agent. The predictive model is trained on training data generated by selectively assigning response templates to agent messages from text dialogues. Examples enable a predictive model to be trained on high quality data sets that are generated automatically from a corpus of historical data. In turn, they enable a natural language interface to be efficiently provided.

Abstract: Certain examples described herein allow feedback to be exchanged between a conversational agent and an operator (so-called “bi-directional” feedback). Certain examples allow an incorrect response template to be indicated by the operator and the conversational agent to compute a contribution for tokens representative of how influential the tokens were in the prediction of the incorrect response template by an applied predictive model. The computed contribution is used to provide further feedback to the operator comprising potential tokens to disassociate with the incorrect response template. The operator then selects the tokens they wish to disassociate and the parameters of the predictive model are adjusted based on this feedback. By repeating this process, an accuracy of a conversational agent, in the form of the response templates that are selectable for a text dialogue, may be improved.

Abstract: Certain examples are described that provide methods and systems for generating templates for use by a conversational agent. These examples enable a natural language interface to be provided. Certain examples cluster user and agent messages from a corpus of text data representing text dialogues. This clustering enables response templates to be generated in a way that takes into account a context in which responses are given. In certain examples, messages that are exchanged between a user and a conversational agent are embedded as numeric arrays based a neural sequence-to-sequence model. Clustering routines are used to group dialogue encodings into one or more response clusters, and these clusters may then be used to generate response templates. The response templates may be used by a conversational agent to prepare a response to a user message.

Abstract: A method of controlling access to content such as web sites on the intranet or interne is described. For example, the method comprises receiving an address of content to be accessed and obtaining similarity of the address to previously labeled addresses of other content items. The similarity is based on co-occurrence of addresses of content items in records of browsing sessions from many consenting users. For example, a browsing session record comprises addresses of content items accessed by a user in a time period during which the user is actively accessing content. A co-occurrence of addresses of content items is the existence of the addresses in the same browsing session record. Access to the content is then controlled on the basis of the similarity.

Abstract: A method comprises: receiving input of a dialogue; processing the dialogue by a neural network based system, to output, for each of a plurality of slots, a probability distribution over a range of values associated with the respective slot, the neural network based system being trained using a training dataset comprising a plurality of dialogues and, for each dialogue, a value corresponding to each slot, wherein each dialogue resulted in an action; determining, based at least on the probability distribution for each slot, if an action requiring one of values for at least some of the slots can be performed; if not, causing continuing of the dialogue.

Abstract: A computer-implemented method of tagging a text, comprises: determining a value for each of a plurality of locations in a first vector; processing (402), by a trained first neural network component, the first vector to generate a second vector; processing (404), at a trained second neural network component, the second vector to generate a probability score for each of at least ten predetermined tags; determining (406) if each probability score meets a criterion; if the criterion is met, assigning (408) the tag corresponding to the probability score to the text. Each of the locations may correspond to a respective predetermined word, each value relating to existence and/or frequency of the corresponding word in the text, and the number of locations may be between 600 and 20000.

Abstract: Online recommendations are tracked through a forwarding service. The forwarding service can provide such statistics to an ad service, which can provide incentives to the recommending user and a consuming user. Example incentives may include an accumulation of points by the recommending user, a discount to the consuming user if a purchase is made in response to the recommendation, etc. To determine how much of an incentive each participant in the recommendation flow receives, a graph is created to model the recommendation flow and incentives are allocated using a cooperative game description based on this graph that associates each participant with a power index that represents that participants share of the incentive.

Abstract: A method of controlling access to content such as web sites on the intranet or interne is described. For example, the method comprises receiving an address of content to be accessed and obtaining similarity of the address to previously labeled addresses of other content items. The similarity is based on co-occurrence of addresses of content items in records of browsing sessions from many consenting users. For example, a browsing session record comprises addresses of content items accessed by a user in a time period during which the user is actively accessing content. A co-occurrence of addresses of content items is the existence of the addresses in the same browsing session record. Access to the content is then controlled on the basis of the similarity.

Abstract: Checking and/or completing for data grids is described such as for grids having rows and columns of cells at least some of which contain data values such as numbers or categories. In various embodiments predictive probability distributions are obtained from an inference engine for one or more of the cells and the predictive probability distributions are used for various tasks such as to suggest values to complete blank cells, highlight cells having outlying values, identify potential errors, suggest corrections to potential errors, identify similarities between cells, identify differences between cells, cluster rows of the data grid, and other tasks. In various embodiments a graphical user interface displays a data grid and provides facilities for completing, error checking/correcting, and analyzing data in the data grid.

Abstract: Data structures for collaborative filtering systems are described. In an embodiment sketches which extremely concisely represent a list of items that a user has rated are created and stored for use by a collaborative filtering system to recommend items. For example, the sketches are created by using several versions of a cryptographic hash function to permute the item list and store a minimal value from each permutation in the sketch together with a user rating. In examples the sketches are used to compute estimates of similarity measures between pairs of users such as rank correlations including Spearman's Rho and Kendall's Tau. For example, the similarity measures are used by a collaborative filtering system to accurately and efficiently recommend items to users. For example the sketches are so concise that massive amounts of data can be taken into account in order to give high quality recommendations in a practical manner.

Abstract: Online recommendations are tracked through a forwarding service. The forwarding service can provide such statistics to an ad service, which can provide incentives to the recommending user and a consuming user. Example incentives may include an accumulation of points by the recommending user, a discount to the consuming user if a purchase is made in response to the recommendation, etc. To determine how much of an incentive each participant in the recommendation flow receives, a graph is created to model the recommendation flow and incentives are allocated using a cooperative game description based on this graph that associates each participant with a power index that represents that participants share of the incentive.