Search Results for author:
Found 25 papers, 12 papers with code
Latent Causal Probing: A Formal Perspective on Probing with Causal Models of Data
As language models (LMs) deliver increasing performance on a range of NLP tasks, probing classifiers have become an indispensable technique in the effort to better understand their inner workings.
Depth-bounded Epistemic Logic
Existing logics typically assume the ability of agents to reason perfectly about propositions of unbounded modal depth.
Sound Explanation for Trustworthy Machine Learning
We take a formal approach to the explainability problem of machine learning systems.
Emergent Representations of Program Semantics in Language Models Trained on Programs
Each program in the corpus is preceded by a (partial) specification in the form of several input-output grid world states.
Effective Neural Network $L_0$ Regularization With BinMask
$L_0$ regularization of neural networks is a fundamental problem.
Defending Against Backdoor Attacks Using Ensembles of Weak Learners
We propose an iterative training procedure for removing poisoned data from the training set.
Efficient Regularization for Adversarially Robustness Deep ReLU Networks
Crucially, our models are simultaneously robust against multiple state-of-the-art adversaries, suggesting that the robustness generalizes well to \textit{unseen} adversaries.
Verifying Low-dimensional Input Neural Networks via Input Quantization
We propose to prepend an input quantization layer to the network.
Incompatibility Clustering as a Defense Against Backdoor Poisoning Attacks
We propose a novel clustering mechanism based on an incompatibility property between subsets of data that emerges during model training.
Inductive Program Synthesis over Noisy Datasets using Abstraction Refinement Based Optimization
Both Rose and the previous system synthesize programs that are optimal over the provided noisy data sets.
Program Synthesis Over Noisy Data with Guarantees
We also formalize the concept and conditions required for convergence, i. e., conditions under which the probability that the synthesis algorithm produces a correct program increases as the size of the noisy data set increases.
Program Synthesis Guided Reinforcement Learning for Partially Observed Environments
Our results demonstrate that our approach can obtain the benefits of program-guided reinforcement learning without requiring the user to provide a new guiding program for every new task.
Neurosymbolic Transformers for Multi-Agent Communication
We study the problem of inferring communication structures that can solve cooperative multi-agent planning problems while minimizing the amount of communication.
Context-Agnostic Learning Using Synthetic Data
We propose a novel setting for learning, where the input domain is the image of a map defined on the product of two sets, one of which completely determines the labels.
Towards Context-Agnostic Learning Using Synthetic Data
We propose a novel setting for learning, where the input domain is the image of a map defined on the product of two sets, one of which completely determines the labels.
Efficient Exact Verification of Binarized Neural Networks
We present a new system, EEV, for efficient and exact verification of BNNs.
Manifold Regularization for Locally Stable Deep Neural Networks
We apply concepts from manifold regularization to develop new regularization techniques for training locally stable deep neural networks.
Exploiting Verified Neural Networks via Floating Point Numerical Error
For a pretrained neural network, we present a method that efficiently searches inputs as witnesses for the incorrectness of robustness claims made by a complete verifier.
Correctness Verification of Neural Networks
The presented framework also enables detecting illegal inputs -- inputs that are not contained in (or close to) the target input space as defined by the state space and observation process (the neural network is not designed to work on them), so that we can flag when we don't have guarantees.
A Hardware Platform for Efficient Multi-Modal Sensing with Adaptive Approximation
We present Warp, a hardware platform to support research in approximate computing, sensor energy optimization, and energy-scavenged systems.
Applied Physics Hardware Architecture Emerging Technologies Robotics Instrumentation and Detectors
Incremental Color Quantization for Color-Vision-Deficient Observers Using Mobile Gaming Data
We introduce DaltonQuant, a new color quantization technique for image compression that cloud services can apply to images destined for a specific user with known color vision deficiencies.
The Three Pillars of Machine Programming
In this position paper, we describe our vision of the future of machine programming through a categorical examination of three pillars of research.
Unanimous Prediction for 100% Precision with Application to Learning Semantic Mappings
Specifically, we introduce the unanimity principle: only predict when all models consistent with the training data predict the same output.
