Abstract: Methods, computer program products, and systems are presented. The method computer program products, and systems can include, for instance: applying a first data stream to a first encoder network that produces a first embedding of the first data stream and inputting the first embedding to a first decoder network that produces a first reconstructed stream of the first data stream; inputting a second data stream to a second encoder network that produces a second embedding of the second data stream and inputting the second embedding to a second decoder network that produces a second reconstructed stream of the second data stream, wherein the first encoder network, the first decoder network, the second encoder network, and the second decoder network define a predictive model; and determining an embedding loss function between the first embedding and the second embedding.

Abstract: Aspects of the invention include systems and methods configured to extract enriched target-oriented common sense from grounded graphs to support efficient next step decision making of an autonomous agent. A non-limiting example computer-implemented method includes extracting common sense from a source. The extracted common sense can include a first knowledge graph. An environment state can be extracted from an observation. The extracted environment state can include a second knowledge graph. The second knowledge graph can include an interactive object and a state of the interactive object. A difference graph including the extracted common sense and the extracted environment state can be generated. A next action is selected based on the difference graph and the next action is taken by an autonomous agent.

Abstract: Self-supervised learning of a machine learning model using images. The computing device masks a section of each image in an image database to generate a partially masked image. An autoencoder encodes each partially masked image to generate one or more encodings representing each partially masked image. The autoencoder decodes each of the one or more encodings into one or more decoded encodings representing each partially masked image previously input into the autoencoder. The computing device compares each of the one or more decoded encodings of each partially masked image with the corresponding image from the image database to generate an unaugmented model output. The computing device augments each image according to a data augmentation policy to generate an augmented model output. The computing device determines a total loss by comparing the unaugmented model output to the augmented model output. The autoencoder is improved based upon the total loss.

Abstract: Mechanisms are provided for answering a multi-hop question. The mechanisms extract one or more entities included in the multi-hop question and generate, for each entity, a plurality of sub-questions to help answer the multi-hop question. The mechanisms obtain an answer to each sub-question from a knowledge base to convert each pair of the answer and the sub-question into each affirmative sentence. The mechanisms generate one or more reasoning sentences to answer the multi-hop question by using one or more affirmative sentences and determine whether the multi-hop question is answerable or not by using the one or more reasoning sentences. The mechanisms, in response to a positive determination, output an answer to the multi-hop question by using the one or more affirmative sentences and the one or more reasoning sentences.

Abstract: Next state prediction technology that performs the following computer based operations: receiving state information that includes information indicative of a current state of an environment; processing the state information to predict a future state of the environment, with the processing being performed by a hybrid computer system that includes both of the following: (i) neural network software module(s) that include machine learning functionality, and (ii) symbolic rule based software modules; and using the prediction of the next state of the environment as an input with respect to taking a further action (for example, activating a hardware device or effecting a communication to a human or another device).

Abstract: An embodiment for generating and employing incrementally optimized combinatorial prompts for tuning a target model. The embodiment may select a predetermined number of examples from a training dataset. The embodiment may concatenate each of the selected examples with a current prompt of a target model to obtain a set of candidate prompts. The embodiment may, for each individual candidate prompt in the set of candidate prompts, calculate a loss value over a validation dataset. The embodiment may replace the current prompt with the individual candidate prompt having a lowest calculated loss value that is less than or equal to an original loss value over the validation set for the current prompt to obtain an updated prompt.

Abstract: A computer-implemented method for reinforcement learning with Logical Neural Networks (LNNs) is provided including receiving a plurality of observation text sentences from a target environment, extracting one or more propositional logic values from the plurality of observation text sentences, finding a class for each propositional logic value by using external knowledge, converting each propositional logic value into a first-order logic by replacing a part in the propositional logic value with a variable word, the part indicating the class, selecting a LNN based on the class among LNNs prepared in advance for each class, each LNN receiving the one or more propositional logic values as a status input and outputting an action with a score indicating a degree of preference for taking the action, and performing a highest score action to the target environment to obtain a next state of the target environment and a reward for the highest score action.

Abstract: A computer-implemented method is provided for learning multimodal feature matching. The method includes training an image encoder to obtain encoded images. The method further includes training a common classifier on the encoded images by using labeled images. The method also includes training a text encoder while keeping the common classifier in a fixed configuration by using learned text embeddings and corresponding labels for the learned text embeddings. The text encoder is further trained to match a distance of predicted text embeddings which is encoded by the text encoder to a fitted Gaussian distribution on the encoded images.

Abstract: A computer-implemented method, according to one embodiment, includes extracting predicates from a predetermined plurality of sentences, and causing an explainer component to analyze the sentences to determine attentions from the predicates of the sentences. The method further includes causing the extracted predicates to be input into a predetermined pruner model. The pruner model is trained to use the attentions to generate a pruned list of predicates from the extracted predicates. A logical neural network is caused to be trained using the pruned list of predicates. A computer program product, according to another embodiment, includes a computer readable storage medium having program instructions embodied therewith. The program instructions are readable and/or executable by a computer to cause the computer to perform the foregoing method.

Abstract: Self-supervised learning of a machine learning model using images. The computing device masks a section of each image in an image database to generate a partially masked image. An autoencoder encodes each partially masked image to generate one or more encodings representing each partially masked image. The autoencoder decodes each of the one or more encodings into one or more decoded encodings representing each partially masked image previously input into the autoencoder. The computing device compares each of the one or more decoded encodings of each partially masked image with the corresponding image from the image database to generate an unaugmented model output. The computing device augments each image according to a data augmentation policy to generate an augmented model output. The computing device determines a total loss by comparing the unaugmented model output to the augmented model output. The autoencoder is improved based upon the total loss.

Abstract: Examples described herein provide a method for explaining neuro-symbolic reinforcement learning reasoning in a neuro-symbolic neural network for neuro-symbolic artificial intelligence. The method includes selecting an action from among possible candidates taken in an environment, wherein the action comprises a pair of a verb and an entity and displaying one or more logical facts that are extracted from natural observation sentences of the environment. The method also includes visualizing contrastive information for a current state and a goal state which is from external knowledge and displaying trained rules in the neuro-symbolic neural network for neuro-symbolic artificial intelligence, wherein, in response to a first user selection of the action, highlighting each pair of the verb and the entity and a fired predicate corresponding to the first user selection.

Abstract: Mechanisms are provided for a model-based Reinforcement Learning (RL) computing system. A proprioception module receives a previous state of an environment and a previous action taken by an agent in the environment, and estimates a current state by using a transition model which receives a pair of state and action and produces a next state. The proprioception module modifies an estimate of the transition model so that the modified estimate of the transition model prevents a past invalid action from recurring in a corresponding state, where the past invalid action taken in the corresponding state is one that did not cause a change in state. The proprioception module passes the current state and the modified estimate of the transition model to a model-based RL computer model for generation of a next action to take in the environment.

Abstract: Aspects of the invention include systems and methods configured to extract enriched target-oriented common sense from grounded graphs to support efficient next step decision making of an autonomous agent. A non-limiting example computer-implemented method includes extracting common sense from a source. The extracted common sense can include a first knowledge graph. An environment state can be extracted from an observation. The extracted environment state can include a second knowledge graph. The second knowledge graph can include an interactive object and a state of the interactive object. A difference graph including the extracted common sense and the extracted environment state can be generated. A next action is selected based on the difference graph and the next action is taken by an autonomous agent.

Abstract: Aspects of the present invention disclose a method for a distance-based vector classification in anomaly detection. The method includes one or more processors identifying one or more audio communications from a first user to a second user, wherein the one or more audio communications is transmitted utilizing a first computing device. The method further includes determining an objective of the first user based at least in part on the audio communication of the first user. The method further includes determining a set of conditions corresponding to the one or more audio communications and the objective, wherein the set of conditions indicate a vulnerability of personal data of the first user. The method further includes prohibiting the first computing device from transmitting audio data that includes the personal data of the first user.

Abstract: A computer-implemented method for reinforcement learning with Logical Neural Networks (LNNs) is provided including receiving a plurality of observation text sentences from a target environment, extracting one or more propositional logic values from the plurality of observation text sentences, finding a class for each propositional logic value by using external knowledge, converting each propositional logic value into a first-order logic by replacing a part in the propositional logic value with a variable word, the part indicating the class, selecting a LNN based on the class among LNNs prepared in advance for each class, each LNN receiving the one or more propositional logic values as a status input and outputting an action with a score indicating a degree of preference for taking the action, and performing a highest score action to the target environment to obtain a next state of the target environment and a reward for the highest score action.

Abstract: In an approach, a processor trains a model, via a reinforcement learning process, to produce a first action function for relating states of a natural language based response environment to actions applicable to the natural language based response environment.

Abstract: A computer-implemented method is provided for learning an action policy. The method includes obtaining, by a processor, environment dynamics including triplets of a state, an action, and a next state. The state in each of the triplets is an expert state. The method further includes training, by the processor using the environment dynamics as training data, a dynamics model which obtains a pair of the state and the action as an input and outputs, for each next state, state-transition probabilities. The method also includes learning, by the processor, the action policy using trajectories of expert states according to a supervised learning technique by back-propagating error gradients through the trained dynamics model.

Abstract: Next state prediction technology that performs the following computer based operations: receiving state information that includes information indicative of a current state of an environment; processing the state information to predict a future state of the environment, with the processing being performed by a hybrid computer system that includes both of the following: (i) neural network software module(s) that include machine learning functionality, and (ii) symbolic rule based software modules; and using the prediction of the next state of the environment as an input with respect to taking a further action (for example, activating a hardware device or effecting a communication to a human or another device).

Abstract: A computer-implemented method is provided for learning multimodal feature matching. The method includes training an image encoder to obtain encoded images. The method further includes training a common classifier on the encoded images by using labeled images. The method also includes training a text encoder while keeping the common classifier in a fixed configuration by using learned text embeddings and corresponding labels for the learned text embeddings. The text encoder is further trained to match a distance of predicted text embeddings which is encoded by the text encoder to a fitted Gaussian distribution on the encoded images.

Abstract: A method for action pruning in Reinforcement Learning receives a current state of an environment. The method evaluates, using a Logical Neural Network (LNN) structure, a logical inference based on the current state. The method outputs upper and lower bounds on each action from a set of possible actions of an agent in the environment, responsive to an evaluation of the logical inference. The method calculates, for each pair of a possible action of the agent in the environment and the current state, a probability by using the upper and lower bounds. Each of calculated probabilities indicates a respective priority ratio for the each action. The method obtains a policy in Reinforcement Learning for the current state by using the calculated probabilities. The method prunes one or more actions from the set of actions as being in violation of the policy such that the one or more actions are ignored.