Abstract: The present disclosure relates to medical and health decision making and, more particularly, to treatment based on tumor clonality estimates. Methods and systems include analyzing genotypes of a tumor to identify clonality sub-types present in the tumor, using a machine learning model that is trained to learn a multilevel evolutionary process or genetic algorithm, by using a recursive Wasserstein objective to output the clonal sub-types, an ancestral structure, and a fitness model. A treatment is generated, tailored to the tumor using the clonal sub-types, and subclonal properties predicted by the model, such as subclone fitness.

Abstract: Methods and systems for three-dimensional (3D) molecule generation include training an autoencoder machine learning model that disentangles structural context of a molecule from properties of the molecule, using a loss function that further enforces equivariance of a coordinate representation and invariance of data likelihood. A 3D molecule is generated using the trained autoencoder machine learning model.

Abstract: A device may convert a motion dataset to a compatible representation for use in a physics engine, wherein the physics engine includes a physics simulator and inverse dynamics network. The device may further downsample the motion dataset to obtain keyframes for motion generation and forming a downsampled motion dataset. The device may further execute a deep generative model based on the downsampled motion dataset to generate a first generated motion. The device may further execute the physics engine by feeding pairs of consecutive keyframes into the physics simulator and the inverse dynamics network to generate a second generated motion. The device may further combine the first generated motion and the second generated motion to form a combined generated motion, wherein the combined generated motion is generated by executing the physics engine with the first generated motion. The device may further generate a simulated motion video from the combined generated motion.

Abstract: Methods and systems for peptide binding prediction include predicting a three-dimensional (3D) structure of a peptide and a major histocompatibility (MHC) complex to generate a graph. The 3D structure is refined by pruning edges of the graph having a distance between the peptide and the MHC complex that is below a threshold value. Models for MHC-I and MHC-II binding prediction are trained, including Bayesian reweighting of data for the MHC-II binding prediction, using the pruned graph.

Abstract: Systems and methods include predicting a first action step and a last action step based on an initial visual observation and a goal visual state and retrieving multiple procedural plans from a procedural knowledge graph (PKG), trained using a set of training instructional videos, which start with the first action step and end with the last action step. A procedure plan is generated using the retrieved multiple procedural plans. An instructional video is generated based on the procedure plan.

Abstract: Systems and methods include collecting real-world distributed-optic fiber sensing (DFOS) sensing data from a target environment as a reference dataset. A synthetic sketch dataset is constructed as a parameterized computer program. A synthetic waterfall is generated from a deep neural network as an image translator from the sketch waterfall with nonlinear distortions and background noises added. Parameters are optimized for generating the synthetic waterfall under a loss function where the loss function encodes a generalization performance on the real-world dataset and encodes granularities from a sensing process and uncontrollable factors.

Abstract: Methods and systems for action detection include encoding a text feature of an input textual description of an action using a visual language model (VLM). A video feature of an input video is encoded using the VLM. The action in the video is recognized, based on the text feature and the video feature, to localize the action within the video. A person performing the action is located within the video using the VLM.

Abstract: A method for learning disentangled representations of videos is presented. The method includes feeding each frame of video data into an encoder to produce a sequence of visual features, passing the sequence of visual features through a deep convolutional network to obtain a posterior of a dynamic latent variable and a posterior of a static latent variable, sampling static and dynamic representations from the posterior of the static latent variable and the posterior of the dynamic latent variable, respectively, concatenating the static and dynamic representations to be fed into a decoder to generate reconstructed sequences, and applying three regularizers to the dynamic and static latent variables to trigger representation disentanglement. To facilitate the disentangled sequential representation learning, orthogonal factorization in generative adversarial network (GAN) latent space is leveraged to pre-train a generator as a decoder in the method.

Abstract: Methods and systems for molecule generation include embedding an input template molecule into a latent space to generate a vector. The vector is decoded using a denoising diffusion implicit model (DDIM) to generate a new molecule specification that is based on the input template molecule. The new molecule is produced using the new molecule specification.

Abstract: Methods and systems for answering a query include generating first tokens in response to an input query using a language model, the first tokens including a retrieval rule. A retrieval rule is used to search for information to generate dynamic tokens. The retrieval rule in the first tokens is replaced with the dynamic tokens to generate a dynamic partial response. Second tokens are generated in response to the input query. The second tokens are appended to the dynamic partial response to generate an output responsive to the input query.

Abstract: A reinforcement learning based approach to the problem of query object localization, where an agent is trained to localize objects of interest specified by a small exemplary set. We learn a transferable reward signal formulated using the exemplary set by ordinal metric learning. It enables test-time policy adaptation to new environments where the reward signals are not readily available, and thus outperforms fine-tuning approaches that are limited to annotated images. In addition, the transferable reward allows repurposing of the trained agent for new tasks, such as annotation refinement, or selective localization from multiple common objects across a set of images. Experiments on corrupted MNIST dataset and CU-Birds dataset demonstrate the effectiveness of our approach.

Abstract: A method for graph analysis includes identifying trainable control parameters of a graph refinement function. Sample graph refinements of an input graph are generated, using control parameters sampled from a variational distribution. Graph refinement control parameters associated with a sample graph refinement that has a highest performance score are selected when used to train a graph neural network. Graph analysis is performed on the input graph using the selected graph refinement parameters to produce a refined graph on new test samples. An action is performed responsive to the graph analysis.

Abstract: A computer-implemented method is provided for disentangled data generation. The method includes accessing, by a bidirectional Long Short-Term Memory (LSTM) with a multi-head attention mechanism, a dataset including a plurality of pairs each formed from a given one of a plurality of input text structures and given one of a plurality of style labels for the plurality of input text structures. The method further includes training the bidirectional LSTM as an encoder to disentangle a sequential text input into disentangled representations comprising a content embedding and a style embedding based on a subset of the dataset. The method also includes training a unidirectional LSTM as a decoder to generate a next text structure prediction for the sequential text input based on previously generated text structure information and a current word, from a disentangled representation with the content embedding and the style embedding.

Abstract: A method for implementing deep reinforcement learning with T-cell receptor (TCR) mutation policies to generate binding TCRs recognizing target peptides for immunotherapy is presented. The method includes extracting peptides to identify a virus or tumor cells, collecting a library of TCRs from target patients, predicting, by a deep neural network, interaction scores between the extracted peptides and the TCRs from the target patients, developing a deep reinforcement learning (DRL) framework with TCR mutation policies to generate TCRs with maximum binding scores, defining reward functions based on a reconstruction-based score and a density estimation-based score, randomly sampling batches of TCRs and following a policy network to mutate the TCRs, outputting mutated TCRs, and ranking the outputted TCRs to utilize top-ranked TCR candidates to target the virus or the tumor cells for immunotherapy.

Abstract: A method for implementing deep reinforcement learning with T-cell receptor (TCR) mutation policies to generate binding TCRs recognizing target peptides for immunotherapy is presented. The method includes extracting peptides to identify a virus or tumor cells, collecting a library of TCRs from target patients, predicting, by a deep neural network, interaction scores between the extracted peptides and the TCRs from the target patients, developing a deep reinforcement learning (DRL) framework with TCR mutation policies to generate TCRs with maximum binding scores, defining reward functions based on a reconstruction-based score and a density estimation-based score, randomly sampling batches of TCRs and following a policy network to mutate the TCRs, outputting mutated TCRs, and ranking the outputted TCRs to utilize top-ranked TCR candidates to target the virus or the tumor cells for immunotherapy.

Abstract: A method for implementing deep reinforcement learning with T-cell receptor (TCR) mutation policies to generate binding TCRs recognizing target peptides for immunotherapy is presented. The method includes extracting peptides to identify a virus or tumor cells, collecting a library of TCRs from target patients, predicting, by a deep neural network, interaction scores between the extracted peptides and the TCRs from the target patients, developing a deep reinforcement learning (DRL) framework with TCR mutation policies to generate TCRs with maximum binding scores, defining reward functions based on a reconstruction-based score and a density estimation-based score, randomly sampling batches of TCRs and following a policy network to mutate the TCRs, outputting mutated TCRs, and ranking the outputted TCRs to utilize top-ranked TCR candidates to target the virus or the tumor cells for immunotherapy.

Abstract: Methods and systems for training a model include performing spatial augmentation on an unlabeled input video to generate spatially augmented video. Temporal augmentation is performed on the input video to generate temporally augmented video. Predictions are generated, using a model that was pre-trained on a labeled dataset, for the unlabeled input video, the spatially augmented video, and the temporally augmented video. Parameters of the model are adapted using the predictions while enforcing temporal consistency, temporal consistency, and historical consistency. The model may be used for action recognition in a healthcare context, with recognition results being used for determining whether patients are performing a rehabilitation exercise correctly.

Abstract: Methods and systems for training a model include training an encoder in an unsupervised fashion based on a backward latent flow between a reference frame and a driving frame taken from a same video. A diffusion model is trained that generates a video sequence responsive to an input image and a text condition, using the trained encoder to determine a latent flow sequence and occlusion map sequence of a labeled training video.

Abstract: A method for generating binding peptides presented by any given Major Histocompatibility Complex (MHC) protein is presented. The method includes, given a peptide and an MHC protein pair, enabling a Reinforcement Learning (RL) agent to interact with and exploit a peptide mutation environment by repeatedly mutating the peptide and observing an observation score of the peptide, learning to form a mutation policy, via a mutation policy network, to iteratively mutate amino acids of the peptide to obtain desired presentation scores, and generating, based on the desired presentation scores, qualified peptides and binding motifs of MHC Class I proteins.

Abstract: Methods and systems for image generation include generating a latent representation of an image, modifying the latent representation of the image based on a trained attribute classifier and a specified attribute input, and decoding the modified latent representation to generate an output image that matches the specified attribute input.