Search Results for author: Johannes Schneider

Found 35 papers, 8 papers with code

Embracing the Enemy

no code implementations14 Jun 2024 Johannes Schneider, Álvaro Delgado-Vega

Even when fully aligned with the friend, the principal embraces the enemy by persistently endorsing him once an initial ``cordon sanitaire'' to exclude the enemy breaks exogenously.

Governance of Generative Artificial Intelligence for Companies

no code implementations5 Feb 2024 Johannes Schneider, Rene Abraham, Christian Meske

Generative Artificial Intelligence (GenAI), specifically large language models like ChatGPT, has swiftly entered organizations without adequate governance, posing both opportunities and risks.

Negotiating with LLMS: Prompt Hacks, Skill Gaps, and Reasoning Deficits

no code implementations26 Nov 2023 Johannes Schneider, Steffi Haag, Leona Chandra Kruse

Large language models LLMs like ChatGPT have reached the 100 Mio user barrier in record time and might increasingly enter all areas of our life leading to a diverse set of interactions between those Artificial Intelligence models and humans.

Improving classifier decision boundaries using nearest neighbors

no code implementations5 Oct 2023 Johannes Schneider

Neural networks are not learning optimal decision boundaries.

Towards LLM-based Autograding for Short Textual Answers

no code implementations9 Sep 2023 Johannes Schneider, Bernd Schenk, Christina Niklaus, Michaelis Vlachos

Thus, in this manuscript we provide an evaluation of a large language model for the purpose of autograding, while also highlighting how LLMs can support educators in validating their grading procedures.

Decision Making Language Modelling +1

SoK: Pragmatic Assessment of Machine Learning for Network Intrusion Detection

1 code implementation30 Apr 2023 Giovanni Apruzzese, Pavel Laskov, Johannes Schneider

Unfortunately, the value of ML for NID depends on a plethora of factors, such as hardware, that are often neglected in scientific literature.

Network Intrusion Detection

Foundation models in brief: A historical, socio-technical focus

no code implementations17 Dec 2022 Johannes Schneider

Foundation models can be disruptive for future AI development by scaling up deep learning in terms of model size and training data's breadth and size.

In-Context Learning

Concept-based Adversarial Attacks: Tricking Humans and Classifiers Alike

no code implementations18 Mar 2022 Johannes Schneider, Giovanni Apruzzese

We propose to generate adversarial samples by modifying activations of upper layers encoding semantically meaningful concepts.

Decision Making

Explaining Classifiers by Constructing Familiar Concepts

1 code implementation7 Mar 2022 Johannes Schneider, Michail Vlachos

Our proposed `CLAssifier-DECoder' architecture (ClaDec) facilitates the understanding of the output of an arbitrary layer of neurons or subsets thereof.

Decoder Image Classification

The learning phases in NN: From Fitting the Majority to Fitting a Few

no code implementations16 Feb 2022 Johannes Schneider

The learning dynamics of deep neural networks are subject to controversy.

Mechanism Design with Informational Punishment

no code implementations4 Jan 2022 Benjamin Balzer, Johannes Schneider

We introduce \emph{informational punishment} to the design of mechanisms that compete with an exogenous status quo mechanism: Players can send garbled public messages with some delay, and others cannot commit to ignoring them.

Towards Trustworthy AutoGrading of Short, Multi-lingual, Multi-type Answers

no code implementations2 Jan 2022 Johannes Schneider, Robin Richner, Micha Riser

We also show how teachers can effectively control the type of errors made by the system and how they can validate efficiently that the autograder's performance on individual exams is close to the expected performance.

Math Vocal Bursts Type Prediction

On Risk and Time Pressure: When to Think and When to Do

no code implementations14 Nov 2021 Christoph Carnehl, Johannes Schneider

We show that the agent may return to the initial idea after having left it in the past to explore alternatives.

The Wrong Kind of Information

no code implementations7 Nov 2021 Aditya Kuvalekar, João Ramos, Johannes Schneider

Agents, some with a bias, decide between undertaking a risky project and a safe alternative based on information about the project's efficiency.

Domain Transformer: Predicting Samples of Unseen, Future Domains

1 code implementation10 Jun 2021 Johannes Schneider

The data distribution commonly evolves over time leading to problems such as concept drift that often decrease classifier performance.

Domain Adaptation

A Quest for Knowledge

no code implementations26 Feb 2021 Christoph Carnehl, Johannes Schneider

Researchers select a question and how intensely to study it.

Decision Making

Creativity of Deep Learning: Conceptualization and Assessment

no code implementations3 Dec 2020 Marcus Basalla, Johannes Schneider, Jan vom Brocke

While the potential of deep learning (DL) for automating simple tasks is already well explored, recent research has started investigating the use of deep learning for creative design, both for complete artifact creation and supporting humans in the creation process.

Reflective-Net: Learning from Explanations

1 code implementation27 Nov 2020 Johannes Schneider, Michalis Vlachos

Humans possess a remarkable capability to make fast, intuitive decisions, but also to self-reflect, i. e., to explain to oneself, and to efficiently learn from explanations by others.

AI Governance for Businesses

no code implementations20 Nov 2020 Johannes Schneider, Rene Abraham, Christian Meske, Jan vom Brocke

Artificial Intelligence (AI) governance regulates the exercise of authority and control over the management of AI.

Ethics Management

Towards AI Forensics: Did the Artificial Intelligence System Do It?

no code implementations27 May 2020 Johannes Schneider, Frank Breitinger

Artificial intelligence (AI) makes decisions impacting our daily lives in an increasingly autonomous manner.


Explaining Neural Networks by Decoding Layer Activations

1 code implementation27 May 2020 Johannes Schneider, Michalis Vlachos

We present a `CLAssifier-DECoder' architecture (\emph{ClaDec}) which facilitates the comprehension of the output of an arbitrary layer in a neural network (NN).

Decoder General Classification +1

Correlated Initialization for Correlated Data

no code implementations9 Mar 2020 Johannes Schneider

Our theoretical analysis quantifies the learning behavior of weights of a single spatial filter.

L2 Regularization

Deceptive AI Explanations: Creation and Detection

no code implementations21 Jan 2020 Johannes Schneider, Christian Meske, Michalis Vlachos

To address this issue, our work investigates how AI models (i. e., deep learning, and existing instruments to increase transparency regarding AI decisions) can be used to create and detect deceptive explanations.

text-classification Text Classification

Human-to-AI Coach: Improving Human Inputs to AI Systems

no code implementations8 Dec 2019 Johannes Schneider

We investigate how inputs of humans can be altered to reduce misinterpretation by the AI system and to improve efficiency of input generation for the human while altered inputs should remain as similar as possible to the original inputs.

Personalization of Deep Learning

no code implementations6 Sep 2019 Johannes Schneider, Michail Vlachos

We discuss training techniques, objectives and metrics toward personalization of deep learning models.

BIG-bench Machine Learning Data Augmentation

Locality-Promoting Representation Learning

no code implementations25 May 2019 Johannes Schneider

This work investigates fundamental questions related to learning features in convolutional neural networks (CNN).

Representation Learning

Topic Modeling based on Keywords and Context

2 code implementations7 Oct 2017 Johannes Schneider

Current topic models often suffer from discovering topics not matching human intuition, unnatural switching of topics within documents and high computational demands.

General Classification Topic Models

Obfuscation using Encryption

no code implementations10 Dec 2016 Johannes Schneider, Thomas Locher

We evaluated our method using more than ten programmers as well as pattern mining across open source code repositories to gain insights of (micro-)coding patterns that are relevant for generating misleading statements.

Cryptography and Security

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