This is your work, valued

Pittsburgh, Pennsylvania

Brandon Trabucco

Expert
@brandontrabucco

AI PhD Student @ CMU, Previous Undergraduate @ Berkeley, Musician

da-fusion. Effective Data Augmentation With Diffusion Models

271

design-bench. Benchmarks for Model-Based Optimization

96

design-baselines. Baselines for Model-Based Optimization

55

lstm-cuda. This is a c++ implementation of an LSTM Neural Network parallelized for a GPU using CUDA

25

nerf. Neural Radiance Fields in PyTorch

9

vq-vae-2. Vector Quantized Latent Variable Models In TF 2.0

9

program-gan. We propose an architecture using the recent Generative Adversarial Networks and Hyper Networks to generate syntactically correct and behaviorally useful python source code from erroneous existing code.

7

mass. A Simple Approach For Visual Room Rearrangement: 3D Mapping & Semantic Search (ICLR 2023)

5

scaled_dot_product_attention. Implements the scaled dot product attention mechamism from the transformer. Vaswani, A. et al. https://arxiv.org/pdf/1706.03762.pdf

5

pixelcnn. Conditional Gated Pixel CNN in TensorFlow 2

4

transformer_capsule_layer. Implements a convolutional capsule layer with a scaled dot product attention mechanism.

4

detailed_captioning. Implements state-of-the-art image captioning algorithms, including the bottom-up top-down attention paper.

4

bvn. Birkhoff-von Neumann Decomposition with a Greedy Birkhoff Heuristic

3

tensorflow-hungarian. Guide for building custom op for TensorFlow

3

insta-dev. Official Development Repo for InSTA

3

constrained_beam_search. A tensorflow implementation of constrained beam search that supports arbitrary constraint definitions.

3

morphing-agents. Collection Of Dynamic Morphology Agents For MuJoCo

3

mineral. A minimalist reinforcement learning package for TensorFlow 2.0

3

glove. An implementation of a loader for the GloVe word embeddings.

2

playground. Deep Reinforcement Learning

2

deepfashion_dataset. TensorFlow build script for the deep fashion dataset, depends on glove embeddings.

2

off-policy. Off-Policy Reinforcement Learning Algorithms

2

hypercomplex-ops. Utility functions for TensorFlow that enable working with differentiable hypercomplex numbers.

2

cider. A pure python cider scorer using numpy, see: https://arxiv.org/abs/1411.5726.

2

spork. CLI For Experiments On Slurm / Singularity

2

probabilistic_vq. Implements probabilistic vector quantization for training discrete VAEs

1

separable_attention. Implements a multi head attention mechanism for images.

1

best_first. Non-Sequential Decoding Strategies for Image Captioning

1

wikipedia_dataset. This is a repository using the Wiki Extractor to build and prepare WIKIPEDIA for use in tensorflow.

1

gesture. A Java based gesture recognition app that uses a Convolutional LSTM Neural Network

1

im2txt_match. This is an augmentation of the Show and Tell model trained with an image caption pairwise discriminator.

1