A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.
Improves AI system's understanding of human inputs by creating better examples.
Paper addresses hypothesis space misspecification in learning from human demonstrations and corrections.
High-dimensional neural network manifolds misalign with human perception, causing adversarial examples.
New method uses explicit human demonstrations to teach missing features in reward learning.
Autonomous agents trained via reinforcement learning present numerous safety concerns: reward hacking, negative side effects, and unsafe exploration, among others. In the context of near-future autonomous agents, operating in environments where humans understand the existing dangers, human involvement in the learning p…
Study reveals AI's spontaneous topic changes in text prediction.
MHVAE learns cross-modality inference inspired by human cognition.
Bayesian EnKF improves sentence comprehension uncertainty modeling.
New deep fusion methods improve human action recognition using depth and inertial sensor data.
Robust OPE framework uses human inputs to improve policy evaluation in changing environments.
New method for explaining dialogue response generation models.
Deep reinforcement learning (DRL) has shown incredible performance in learning various tasks to the human level. However, unlike human perception, current DRL models connect the entire low-level sensory input to the state-action values rather than exploiting the relationship between and among entities that constitute t…
When might human input help (or not) when assessing risk in fairness domains? Dressel and Farid (2018) asked Mechanical Turk workers to evaluate a subset of defendants in the ProPublica COMPAS data for risk of recidivism, and concluded that COMPAS predictions were no more accurate or fair than predictions made by human…
Developers of text-to-speech synthesizers (TTS) often make use of human raters to assess the quality of synthesized speech. We demonstrate that we can model human raters' mean opinion scores (MOS) of synthesized speech using a deep recurrent neural network whose inputs consist solely of a raw waveform. Our best models …
We present an approach to interactive-predictive neural machine translation that attempts to reduce human effort from three directions: Firstly, instead of requiring humans to select, correct, or delete segments, we employ the idea of learning from human reinforcements in form of judgments on the quality of partial tra…
Deep neural networks (DNN) trained in a supervised way suffer from two known problems. First, the minima of the objective function used in learning correspond to data points (also known as rubbish examples or fooling images) that lack semantic similarity with the training data. Second, a clean input can be changed by a…
Deep learning model classifies concurrent human interactions from WiFi data with high accuracy.
As machine learning systems move from computer-science laboratories into the open world, their accountability becomes a high priority problem. Accountability requires deep understanding of system behavior and its failures. Current evaluation methods such as single-score error metrics and confusion matrices provide aggr…
CNNs overinterpret inputs, leading to high accuracy without meaningful features.
Paper proposes using pairwise feature comparisons to infer modification costs for user recourse.
Recent studies in the field of human vision science suggest that the human responses to the stimuli on a visual display are non-deterministic. People may attend to different locations on the same visual input at the same time. Based on this knowledge, we propose a new stochastic model of visual attention by introducing…
Training deep reinforcement learning agents complex behaviors in 3D virtual environments requires significant computational resources. This is especially true in environments with high degrees of aliasing, where many states share nearly identical visual features. Minecraft is an exemplar of such an environment. We hypo…
Optimal allocation of human effort to correct AI assessments in decision-making.
The paper fine-tunes language models using human preferences for natural language tasks.
Study finds visual explanations do not significantly improve human accuracy or trust in model predictions.
User interfaces provide an interactive window between physical and virtual environments. A new concept in the field of human-computer interaction is a soft user interface; a compliant surface that facilitates touch interaction through deformation. Despite the potential of these interfaces, they currently lack a signal …
Study shows current metrics for audio adversarial examples are unreliable for human perception.
Develops a statistical framework to measure uncertainty in model rankings based on human preferences.
We present a program synthesis-oriented dataset consisting of human written problem statements and solutions for these problems. The problem statements were collected via crowdsourcing and the program solutions were extracted from human-written solutions in programming competitions, accompanied by input/output examples…
Interpretable companion model for black-box classifiers.
Learning robot objective functions from human input has become increasingly important, but state-of-the-art techniques assume that the human's desired objective lies within the robot's hypothesis space. When this is not true, even methods that keep track of uncertainty over the objective fail because they reason about …
Recent years have seen a boom in interest in machine learning systems that can provide a human-understandable rationale for their predictions or decisions. However, exactly what kinds of explanation are truly human-interpretable remains poorly understood. This work advances our understanding of what makes explanations …
An intriguing property of deep neural networks is their inherent vulnerability to adversarial inputs, which significantly hinders their application in security-critical domains. Most existing detection methods attempt to use carefully engineered patterns to distinguish adversarial inputs from their genuine counterparts…
Framework uses human judgment to distinguish algorithmically indistinguishable cases.
Deep network learns Obstacle Tower challenge without human demonstrations.
Artificial intelligence (AI) is intrinsically data-driven. It calls for the application of statistical concepts through human-machine collaboration during generation of data, development of algorithms, and evaluation of results. This paper discusses how such human-machine collaboration can be approached through the sta…
Few-shot learning is a technique to learn a model with a very small amount of labeled training data by transferring knowledge from relevant tasks. In this paper, we propose a few-shot learning method for wearable sensor based human activity recognition, a technique that seeks high-level human activity knowledge from lo…
Human matting, high quality extraction of humans from natural images, is crucial for a wide variety of applications. Since the matting problem is severely under-constrained, most previous methods require user interactions to take user designated trimaps or scribbles as constraints. This user-in-the-loop nature makes th…
Human analysts that use anomaly detection systems in practice want to retain the use of simple and explainable global anomaly detectors. In this paper, we propose a novel human-in-the-loop learning algorithm called GLAD (GLocalized Anomaly Detection) that supports global anomaly detectors. GLAD automatically learns the…
DreamerV2 learns Atari game behaviors from a world model, achieving human-level performance.
A Human-in-the-Loop Bayesian Optimization framework for constraint-aware bioprocess development.
Tensor neural network improves human pose classification from 3D skeleton data.
CoExBO optimizes lithium-ion batteries with user input, enhancing trust and efficiency.
Recent trends focusing on Industry 4.0 concept and smart manufacturing arise a data-driven fault diagnosis as key topic in condition-based maintenance. Fault diagnosis is considered as an essential task in rotary machinery since possibility of an early detection and diagnosis of the faulty condition can save both time …
This research analyzes how input and output layers affect deep neural networks' resistance to adversarial attacks.
Convolutional Neural Networks (CNNs) have achieved state-of-the-art performance on a variety of computer vision tasks, particularly visual classification problems, where new algorithms reported to achieve or even surpass the human performance. In this paper, we examine whether CNNs are capable of learning the semantics…
Mixed likelihood GPs improve model performance in human-in-the-loop experiments.
While deep learning is remarkably successful on perceptual tasks, it was also shown to be vulnerable to adversarial perturbations of the input. These perturbations denote noise added to the input that was generated specifically to fool the system while being quasi-imperceptible for humans. More severely, there even exi…