Abstract: A computer system that customizes an output provided by a pretrained neural network is described. During operation, the computer system may receive a prompt (or input) associated with a first individual. Then, the computer system may access stored information (in memory) associated with a persona associated with a second individual. For example, the second individual may be different from the first individual. However, in some embodiments, the second individual may be the same as the first individual. Note that the stored information may include memories associated with one or more different sessions of interacting with the pretrained neural network than a current session. Moreover, the computer system may modify the prompt based at least in part on the persona associated with the second individual. Next, the computer system may generate, using the pretrained neural network, the output based at least in part on the modified prompt.

Abstract: A system receives information associated with an interaction with an individual in a context. Then, the system analyzes the information to extract features associated with one or more attributes of the individual. Moreover, the system generates, based at least in part on the extracted features, a group of behavioral agents in a multi-layer hierarchy that automatically mimics the one or more attributes. Next, the system calculates one or more performance metrics associated with the group of behavioral agents and the one or more attributes. Furthermore, the system determines, based at least in part on the one or more performance metrics, one or more deficiencies in the extracted features. Additionally, the system selectively acquires second information associated with additional interaction with the individual in the context based at least in part on the one or more deficiencies to at least in part correct for the one or more deficiencies.

Abstract: A computer system that classifies audio content is described. During operation, the computer system may receive audio content. Then, the computer system may determine a representation of the audio content (such as a signal-processing representation) by performing a transformation on the audio content. In some embodiments, the transformation may include a neural network and/or the representation may include word embedding or sense embedding of words in the audio content. Moreover, the computer system may analyze the representation using a predetermined neural network. Next, the computer system may classify, based at least in part on an output of the predetermined neural network, the audio content as being fake or real, where the fake audio content is, at least in part, computer-generated. Furthermore, the computer system may selectively perform a remedial action based at least in part on the classification.

Abstract: A computer system (which includes one or more computers) that generates a second autoencoder (AE) neural network (such as an ALAP-AE neural network) is described. During operation, the computer system may obtain information specifying an initial AE neural network. Then, the computer system may compute a subset of filters associated with the initial AE neural network to remove based at least in part on a L1-norm loss function and weights associated with filters in initial AE neural network. Moreover, the computer system may prune the subset of the filters from the initial AE neural network. Next, the computer system may generate the ALAP-AE neural network by retraining the initial AE neural network, where the retraining includes a student-teacher model in which the teacher includes the pruned initial AE neural network and the student includes the ALAP-AE neural network.

Abstract: A computer that identifies a fake image is described. During operation, the computer receives an image. Then, the computer performs analysis on the image to determine a signature that includes multiple features. Based at least in part in the determined signature, the computer classifies the image as having a first signature associated with the fake image or as having a second signature associated with a real image, where the first signature corresponds to a finite resolution of a neural network that generated the fake image, a finite number of parameters in the neural network that generated the fake image, or both. For example, the finite resolution may correspond to floating point operations in the neural network. Moreover, in response to the classification, the computer may perform a remedial action, such as providing a warning or a recommendation, or performing filtering.

Abstract: A computer system that trains a neural network is described. During operation, the computer system may receive information specifying a new attack vector corresponding to fake audio content. In response, the computer system may generate a synthetic training dataset based at least in part on the new attack vector. Then, the computer system may access a predetermined neural network that classifies real audio content and fake audio content, where the predetermined neural network was training without synthetic audio content corresponding to the new attack vector. Next, the computer system may train the neural network based at least in part on the synthetic training dataset and the predetermined neural network, where the training of the neural network may include modifying predetermined weights associated with the predetermined neural network, and where a training time for training the neural network may be less than a training time for training the predetermined neural network.

Abstract: An electronic device that provides a virtual representation is described. During operation, the electronic device may receive sensory inputs associated with a user. Then, the electronic device may determine one or more behavioral or emotional inputs based at least in part on the sensory inputs. Moreover, the electronic device may compute a modification to a behavioral or an emotional response of a virtual representation based at least in part on the one or more behavioral or emotional inputs, a behavioral and emotional model, and a set of predefined or predetermined parameters, where the behavioral and emotional model includes one or more purposes that reflect a personality of the virtual representation, and where a given purpose models an expected reward associated with a given need of the virtual representation. Next, the electronic device may provide or present the modification to the behavioral or the emotional response of the virtual representation.

Abstract: A computer system that generates output speech is described. During operation, the computer system may receive an input associated with a type of interaction. Then, the computer system may generate, using a voice synthesis engine, the output speech corresponding to an individual based at least in part on the input, where the voice synthesis engine predicts positions and duration of a prosodic characteristic of speech by the individual, and selectively adds the prosodic characteristic of the speech by the individual in the output speech based at least in part on the prediction. Note that the prosodic characteristic may include: pauses in the speech by the individual, and/or disfluences in the speech by the individual.

Abstract: A computer system that classifies audio content is described. During operation, the computer system may receive audio content. Then, the computer system may determine a representation of the audio content (such as a signal-processing representation) by performing a transformation on the audio content. In some embodiments, the transformation may include a neural network and/or the representation may include word embedding or sense embedding of words in the audio content. Moreover, the computer system may analyze the representation using a predetermined neural network. Next, the computer system may classify, based at least in part on an output of the predetermined neural network, the audio content as being fake or real, where the fake audio content is, at least in part, computer-generated. Furthermore, the computer system may selectively perform a remedial action based at least in part on the classification.

Abstract: A computer system that classifies audio content is described. During operation, the computer system may receive audio content. Then, the computer system may determine a representation of the audio content (such as a signal-processing representation) by performing a transformation on the audio content. In some embodiments, the transformation may include a neural network and/or the representation may include word embedding or sense embedding of words in the audio content. Moreover, the computer system may analyze the representation using a predetermined neural network. Next, the computer system may classify, based at least in part on an output of the predetermined neural network, the audio content as being fake or real, where the fake audio content is, at least in part, computer-generated. Furthermore, the computer system may selectively perform a remedial action based at least in part on the classification.

Abstract: A computer that identifies a fake image is described. During operation, the computer receives an image. Then, the computer performs analysis on the image to determine a signature that includes multiple features. Based at least in part in the determined signature, the computer classifies the image as having a first signature associated with the fake image or as having a second signature associated with a real image, where the first signature corresponds to a finite resolution of a neural network that generated the fake image, a finite number of parameters in the neural network that generated the fake image, or both. For example, the finite resolution may correspond to floating point operations in the neural network. Moreover, in response to the classification, the computer may perform a remedial action, such as providing a warning or a recommendation, or performing filtering.

Abstract: A computer system that generates output speech is described. During operation, the computer system may receive an input associated with a type of interaction. Then, the computer system may generate, using a voice synthesis engine, the output speech corresponding to an individual based at least in part on the input, where the voice synthesis engine predicts positions and duration of a prosodic characteristic of speech by the individual, and selectively adds the prosodic characteristic of the speech by the individual in the output speech based at least in part on the prediction. Note that the prosodic characteristic may include: pauses in the speech by the individual, and/or disfluences in the speech by the individual.

Abstract: A computer system that trains a neural network is described. During operation, the computer system may receive information specifying a new attack vector corresponding to fake audio content. In response, the computer system may generate a synthetic training dataset based at least in part on the new attack vector. Then, the computer system may access a predetermined neural network that classifies real audio content and fake audio content, where the predetermined neural network was training without synthetic audio content corresponding to the new attack vector. Next, the computer system may train the neural network based at least in part on the synthetic training dataset and the predetermined neural network, where the training of the neural network may include modifying predetermined weights associated with the predetermined neural network, and where a training time for training the neural network may be less than a training time for training the predetermined neural network.

Abstract: A computer system that classifies audio content is described. During operation, the computer system may receive audio content. Then, the computer system may determine a representation of the audio content (such as a signal-processing representation) by performing a transformation on the audio content. In some embodiments, the transformation may include a neural network and/or the representation may include word embedding or sense embedding of words in the audio content. Moreover, the computer system may analyze the representation using a predetermined neural network. Next, the computer system may classify, based at least in part on an output of the predetermined neural network, the audio content as being fake or real, where the fake audio content is, at least in part, computer-generated. Furthermore, the computer system may selectively perform a remedial action based at least in part on the classification.

Abstract: A system provides, based at least in part on predetermined parameters, configuration information, and a group of behavioral agents, a dynamic virtual representation that includes a multi-dimensional puppet having one or more attributes of an individual, where the dynamic virtual representation automatically mimics one or more attributes of the individual in a context, the providing of the dynamic virtual representation that includes the multi-dimensional puppet involves rendering of the multi-dimensional puppet, and the multi-dimensional puppet includes stereopsis information, and has photorealistic movement corresponding to movement behaviors of the individual. Then, the system receives an input corresponding to user spatial manipulation of or interaction with the multi-dimensional puppet.

Abstract: A computer that identifies a fake image is described. During operation, the computer receives an image. Then, the computer performs analysis on the image to determine a signature that includes multiple features. Based at least in part in the determined signature, the computer classifies the image as having a first signature associated with the fake image or as having a second signature associated with a real image, where the first signature corresponds to a finite resolution of a neural network that generated the fake image, a finite number of parameters in the neural network that generated the fake image, or both. For example, the finite resolution may correspond to floating point operations in the neural network. Moreover, in response to the classification, the computer may perform a remedial action, such as providing a warning or a recommendation, or performing filtering.

Abstract: A system receives information associated with an interaction with an individual in a context. Then, the system analyzes the information to extract features associated with one or more attributes of the individual. Moreover, the system generates, based at least in part on the extracted features, a group of behavioral agents in a multi-layer hierarchy that automatically mimics the one or more attributes. Next, the system calculates one or more performance metrics associated with the group of behavioral agents and the one or more attributes. Furthermore, the system determines, based at least in part on the one or more performance metrics, one or more deficiencies in the extracted features. Additionally, the system selectively acquires second information associated with additional interaction with the individual in the context based at least in part on the one or more deficiencies to at least in part correct for the one or more deficiencies.

Abstract: A system configures a group of behavioral agents, where a given behavioral agent receives one or more inputs and provides an output corresponding to one or more features associated with an individual, and where the inputs to at least some of the behavioral agents include outputs from one or more of the other behavioral agents. Then, the system generates, based at least in part on a subset of the outputs, a dynamic virtual representation of the individual, where the dynamic virtual representation automatically mimics one or more attributes of the individual in a context. Moreover, the system provides information corresponding to the dynamic virtual representation, and the computer system receives input stimuli associated with a reaction of a user. Next, the system selectively performs a remedial action associated with the group of behavioral agents based at least in part on a portion of the input stimuli.

Abstract: A system provides, based at least in part on predetermined parameters, configuration information, and a group of behavioral agents, a dynamic virtual representation that includes a multi-dimensional puppet having one or more attributes of an individual, where the dynamic virtual representation automatically mimics one or more attributes of the individual in a context, the providing of the dynamic virtual representation that includes the multi-dimensional puppet involves rendering of the multi-dimensional puppet, and the multi-dimensional puppet includes stereopsis information, and has photorealistic movement corresponding to movement behaviors of the individual. Then, the system receives an input corresponding to user spatial manipulation of or interaction with the multi-dimensional puppet.