Seminar

Feature Importance Explanations for Temporal Black-Box Models
Feature Importance Explanations for Temporal Black-Box Models

TIME is a great tool for explaining the predictions made by a model. It is very similar to Grad-CAM, however, it can work with tabular data as well. Also, TIME can provide global expainability as well.

Jun 14, 2023
Entropy-based Logic Explanations of Neural Networks
Entropy-based Logic Explanations of Neural Networks

This research paper aims at building an entropy layer for end-to-end explainable AI framework. Unlike LIME and Grad-CAM, Entropy Net does not rely on any auxuliary tool/model to explain its predictions. Also, Entropy Net focuses on downsizing the concepts required to explain predictions.

May 4, 2023
Rep-Net: Efficient On-Device Learning via Feature Reprogramming
Rep-Net: Efficient On-Device Learning via Feature Reprogramming

REP-Net is a counter to traditional transfer learning for On-board model training with more focus on memory efficiency.

Apr 6, 2023
Training Generative Adversarial Networks with Limited Data
Training Generative Adversarial Networks with Limited Data

Training GANs require large amount of data. If tried with small datasets, the discriminator often times overfit producing meaningless feedback to the generator. One solution to training GANs with smaller dta could be using adaptive data augmentation.

Feb 23, 2023
Domain Knowledge Aided Explainable Artificial Intelligence for Intrusion Detection and Response
Domain Knowledge Aided Explainable Artificial Intelligence for Intrusion Detection and Response

Including domain knowledge can increase explainability without compromising on model performance. This paper mainly discusses a preprocessing technique to incorporate domain knowledge in a way they become very useful for explaining model's predictions.

Feb 1, 2023
Augmented Experiment in Material Engineering Using Machine Learning
Augmented Experiment in Material Engineering Using Machine Learning

Incorporating domain knowledge to neural networks is a creative and case specific approach. This paper modifies the loss function of a fully-connected network with domain knowledge from kinetics which helped the model make precise prediction in its regression task.

Dec 21, 2022
Computational Framework for Sequential Diet Recommendation: Integrating Linear Optimization and Clinical Domain Knowledge
Computational Framework for Sequential Diet Recommendation: Integrating Linear Optimization and Clinical Domain Knowledge

Optimizes change in consumed nutrients while driving users to their desired diet.

Nov 14, 2022
Local Interpretable Model-Agnostic Explanations
Local Interpretable Model-Agnostic Explanations

LIME is a great tool for explaining the predictions made by a model. LIME can explain any model regardless of their type, it works by building a linear model on vicinity of the sample intended to be explained.

Oct 19, 2022
Characterizing Decision Boundary for DNN on High Dimensional Data
Characterizing Decision Boundary for DNN on High Dimensional Data

Decision boundaries are imposible to be visualized in hogh dimensional feature sets. Instead of visualizing them, we can characterize them and make them useful.

Sep 7, 2022
Grad-CAM for Interpreting DNN Model Decisions
Grad-CAM for Interpreting DNN Model Decisions

Neural networks have a bad reputations as they are treated like black boxes and lack interpretations on the results they make. Grad-CAM can slightly interprete what is driving the model to make a decision.

Jul 22, 2022