Search Results for author: Amarda Shehu

Found 10 papers, 2 papers with code

Beyond Single-Model Views for Deep Learning: Optimization versus Generalizability of Stochastic Optimization Algorithms

no code implementations1 Mar 2024 Toki Tahmid Inan, Mingrui Liu, Amarda Shehu

Our investigation encompasses a wide array of techniques, including SGD and its variants, flat-minima optimizers, and new algorithms we propose under the Basin Hopping framework.

Benchmarking Stochastic Optimization

Multiple Instance Learning for Detecting Anomalies over Sequential Real-World Datasets

no code implementations4 Oct 2022 Parastoo Kamranfar, David Lattanzi, Amarda Shehu, Daniel Barbará

The MIL-based formulation performs no worse than single instance learning on easy to moderate datasets and outperforms single-instance learning on more challenging datasets.

Anomaly Detection Multiple Instance Learning

Multi-objective Deep Data Generation with Correlated Property Control

no code implementations1 Oct 2022 Shiyu Wang, Xiaojie Guo, Xuanyang Lin, Bo Pan, Yuanqi Du, Yinkai Wang, Yanfang Ye, Ashley Ann Petersen, Austin Leitgeb, Saleh AlKhalifa, Kevin Minbiole, William Wuest, Amarda Shehu, Liang Zhao

Developing deep generative models has been an emerging field due to the ability to model and generate complex data for various purposes, such as image synthesis and molecular design.

Image Generation

Transformer Neural Networks Attending to Both Sequence and Structure for Protein Prediction Tasks

no code implementations17 Jun 2022 Anowarul Kabir, Amarda Shehu

The increasing number of protein sequences decoded from genomes is opening up new avenues of research on linking protein sequence to function with transformer neural networks.

Interpretable Molecular Graph Generation via Monotonic Constraints

no code implementations28 Feb 2022 Yuanqi Du, Xiaojie Guo, Amarda Shehu, Liang Zhao

Recent advances in deep graph generative models treat molecule design as graph generation problems which provide new opportunities toward the breakthrough of this long-lasting problem.

Disentanglement Drug Discovery +2

Traffic Flow Forecasting with Maintenance Downtime via Multi-Channel Attention-Based Spatio-Temporal Graph Convolutional Networks

no code implementations4 Oct 2021 Yuanjie Lu, Parastoo Kamranfar, David Lattanzi, Amarda Shehu

However, a key shortcoming of state-of-the-art methods is their inability to take into account information of various modalities, for instance the impact of maintenance downtime on traffic flows.


Space Partitioning and Regression Mode Seeking via a Mean-Shift-Inspired Algorithm

no code implementations20 Apr 2021 Wanli Qiao, Amarda Shehu

The mean shift (MS) algorithm is a nonparametric method used to cluster sample points and find the local modes of kernel density estimates, using an idea based on iterative gradient ascent.


Decoy Selection for Protein Structure Prediction Via Extreme Gradient Boosting and Ranking

no code implementations3 Oct 2020 Nasrin Akhter, Gopinath Chennupati, Hristo Djidjev, Amarda Shehu

Consensus methods show varied success in handling the challenge of decoy selection despite some issues associated with clustering large decoy sets and decoy sets that do not show much structural similarity.

BIG-bench Machine Learning Clustering +1

Interpretable Deep Graph Generation with Node-Edge Co-Disentanglement

1 code implementation9 Jun 2020 Xiaojie Guo, Liang Zhao, Zhao Qin, Lingfei Wu, Amarda Shehu, Yanfang Ye

Disentangled representation learning has recently attracted a significant amount of attention, particularly in the field of image representation learning.

Disentanglement Graph Generation

Generating Tertiary Protein Structures via an Interpretative Variational Autoencoder

1 code implementation8 Apr 2020 Xiaojie Guo, Yuanqi Du, Sivani Tadepalli, Liang Zhao, Amarda Shehu

Much scientific enquiry across disciplines is founded upon a mechanistic treatment of dynamic systems that ties form to function.

Protein Structure Prediction Stochastic Optimization

Cannot find the paper you are looking for? You can Submit a new open access paper.