Abstract: Systems and methods for narrative generation. Data representations in a first data domain are received. The data representations of the first data domain are transformed into data representations of a second data domain by an adapter. Narratives corresponding to the data representations can be generated using a large language model by interpreting the data representations of the second data domain.

Abstract: Methods and systems are described for selecting a group of data points to be labelled in an active learning process. The method comprises the step of determining a temperature parameter that selects between a representation-based point selection method and an uncertainty-based point selection method based on the current budget. The method also comprises the step of determining a kernel radius parameter that is inversely proportional to the size of the current budget. The method then includes the step of estimating an uncertainty coverage based on the temperature parameter, the kernel radius parameter and the current budget, wherein the uncertainty coverage is a measure of how much uncertainty is covered by the group of data points. Finally, the next point is greedily selected based on the estimated uncertainty coverage.

Abstract: A fast long-context attention mechanism can be used with any trained transformer model. The best context ranges of tokens for the attention mechanism can be dynamically selected from segments formed from the tokens.

Abstract: A method for training a machine learning engine for event sequence tasks comprises pre-training the machine learning engine using unsupervised learning on at least one pretext task using pretext training data to obtain a partially trained machine learning engine, where the pretext training data comprises pretext event sequences. The method may further comprise, after pre-training the machine learning engine to obtain the partially trained machine learning engine, further training the partially trained machine learning engine on a target task using target task training data to obtain a task-trained machine learning engine, where the target task training data comprises target task event sequences.

Abstract: Methods, systems, and techniques for neural network temporal domain generalization involve training a backbone neural network using a combination of source domains, determining a domain-specific prompt for each of the source domains while the backbone network is frozen, and sequentially determining i) temporal prompts and ii) a general prompt, while training a temporal prompt generator neural network and keeping the backbone network frozen. The various source domains are indexed temporally and respectively are made of data having a time-dependent distribution shift. The temporal prompts capture the dynamics associated with temporal drift in the data, while the general prompt captures general information across all the source domains. This allows the backbone neural network to be adapted to different time periods.

Abstract: Broadly speaking, the present disclosure describes a method of modeling a temporal point process (TPP). For each sequence in the TPP, the method treats the sequence as one of a plurality of distinct tasks, and applies meta learning to the distinct tasks. Applying meta learning to the distinct tasks may comprise applying a neural process to the distinct tasks; the neural process may be an attentive neural process.

Abstract: A neural network for creating representations of time-series may be trained using a self-supervised approach and as such does not require explicit labelling of the training data. The training uses similarity distillation along both the temporal and instance dimensions. Once trained, the neural network may be used to generate representations of a time-series suitable for use on various downstream tasks.

Abstract: Binary classification models can be trained to classify data as being in one of two classes. Membership in a class may be imbalanced so that there are more members in one class than the other. Additionally, one of the classes may have a higher importance than the other, yet appear much less frequently. It is possible to train the binary classification model using a base loss function and a regularization function based on a ranking of training results in order to reduce the false positives at a high true positive rate.

Abstract: A method is provided for training a selective network that includes a selection node for selecting whether to make a prediction. During training, the selection node is reparameterized as a differentiable function of learnable parameters acting on noise from a base distribution. The differentiable function approximates a sampling from a categorical distribution.

Abstract: An artificial neural network for data imbalanced regression and a method for training that network. A regression dataset is obtained that includes multiple pairs that respectively are made up of inputs and corresponding targets. The inputs are represented in a feature space and the targets are represented in a label space of continuous values. Label space similarities between the targets as represented in the label space are determined, and analogously feature space similarities between the inputs as represented in the feature space are determined. A loss may then be determined based on differences between rankings of the label space similarities and corresponding feature space similarities. That loss may be used to train an artificial neural network.

Abstract: A system for machine learning architecture for prospective resource allocations. The system may include a processor and a memory.

Abstract: Systems and methods for knowledge distillation provide supervised training of a student network with a teacher network, including inputting a batch to the teacher network, inputting the batch to the student network, generating a teacher activation map at a layer of the teacher network, generating a student activation map at a layer of the student network corresponding to the layer of the teacher network, generating a pairwise teacher similarity matrix based on the teacher activation map, generating a pairwise student similarity matrix based on the student activation map, and minimizing a knowledge distillation loss defined as a difference between the pairwise teacher similarity matrix and the pairwise student similarity matrix.

Abstract: Disclosed are systems, methods, and devices for computing an action for an automated agent. A neural network configured for deep multi-task learning is provided. Each of a subset of layers of the neural network is connected with a respective gating unit configured for dynamically activating or deactivating the respective layer of the neural network. The method includes: receiving, via a communication interface, input data associated with a task type; selecting, from a plurality of layers of a neural network, a subset of layers based on at least the task type; dynamically activating, based on the input data, at least one layer of the subset of layers; and generating an action signal based on a forward pass of the neural network using the dynamically activated at least one layer of the neural network.

Abstract: A computer-implemented system and method for training a machine learning model are disclosed, the method includes: maintaining a data set representing a neural network having a plurality of weights; receiving input data comprising a plurality of time series data sets ending with timestamp t?1; generating, using the neural network and based on the input data, a probabilistic forecast distribution prediction at timestamp t and a selection value associated with the probabilistic forecast distribution prediction at timestamp t; computing a loss function based on the selection value; and updating at least one of the plurality of weights of the neural network based on the loss function.

Abstract: A computer system and method for training a heterogeneous multi-task learning network is provided. The system comprises at least one processor and a memory storing instructions which when executed by the processor configure the processor to perform the method. The method comprises assigning expert models to each task, processing training input for each task, and storing a final set of weights. For each task, weights in the expert models and in gate parameters are initialized, training inputs are provided to the network, a loss is determined following a forward pass over the network, and losses are back propagated and weights are updated for the experts and the gates. At least one task is assigned one exclusive expert model and at least one shared expert model accessible by the plurality of tasks.

Abstract: Systems and methods for knowledge distillation provide supervised training of a student network with a teacher network, including inputting a batch to the teacher network, inputting the batch to the student network, generating a teacher activation map at a layer of the teacher network, generating a student activation map at a layer of the student network corresponding to the layer of the teacher network, generating a pairwise teacher similarity matrix based on the teacher activation map, generating a pairwise student similarity matrix based on the student activation map, and minimizing a knowledge distillation loss defined as a difference between the pairwise teacher similarity matrix and the pairwise student similarity matrix.