Search Results for author:
Found 123 papers, 31 papers with code
Improving machine classification using human uncertainty measurements
As deep CNN classifier performance using ground-truth labels has begun to asymptote at near-perfect levels, a key aim for the field is to extend training paradigms to capture further useful structure in natural image data and improve model robustness and generalization.
Embodied LLM Agents Learn to Cooperate in Organized Teams
Large Language Models (LLMs) have emerged as integral tools for reasoning, planning, and decision-making, drawing upon their extensive world knowledge and proficiency in language-related tasks.
Learning with Language-Guided State Abstractions
We describe a framework for using natural language to design state abstractions for imitation learning.
Program-Based Strategy Induction for Reinforcement Learning
Typical models of learning assume incremental estimation of continuously-varying decision variables like expected rewards.
Analyzing the Roles of Language and Vision in Learning from Limited Data
Does language help make sense of the visual world?
How do Large Language Models Navigate Conflicts between Honesty and Helpfulness?
In day-to-day communication, people often approximate the truth - for example, rounding the time or omitting details - in order to be maximally helpful to the listener.
A Rational Analysis of the Speech-to-Song Illusion
Here we provide a formal account of this phenomenon, by recasting it as a statistical inference whereby a rational agent attempts to decide whether a sequence of utterances is more likely to have been produced in a song or speech.
Distilling Symbolic Priors for Concept Learning into Neural Networks
Is it possible to create a neural network that displays the same inductive biases?
Human-Like Geometric Abstraction in Large Pre-trained Neural Networks
Humans possess a remarkable capacity to recognize and manipulate abstract structure, which is especially apparent in the domain of geometry.
Measuring Implicit Bias in Explicitly Unbiased Large Language Models
Large language models (LLMs) can pass explicit bias tests but still harbor implicit biases, similar to humans who endorse egalitarian beliefs yet exhibit subtle biases.
Comparing Abstraction in Humans and Large Language Models Using Multimodal Serial Reproduction
To investigate the effect language on the formation of abstractions, we implement a novel multimodal serial reproduction framework by asking people who receive a visual stimulus to reproduce it in a linguistic format, and vice versa.
Preference-Conditioned Language-Guided Abstraction
We observe that how humans behave reveals how they see the world.
Incoherent Probability Judgments in Large Language Models
Autoregressive Large Language Models (LLMs) trained for next-word prediction have demonstrated remarkable proficiency at producing coherent text.
Recovering Mental Representations from Large Language Models with Markov Chain Monte Carlo
Simulating sampling algorithms with people has proven a useful method for efficiently probing and understanding their mental representations.
Concept Alignment
Discussion of AI alignment (alignment between humans and AI systems) has focused on value alignment, broadly referring to creating AI systems that share human values.
Learning Human-like Representations to Enable Learning Human Values
We propose that this kind of representational alignment between machine learning (ML) models and humans can also support value alignment, allowing ML systems to conform to human values and societal norms.
Deep de Finetti: Recovering Topic Distributions from Large Language Models
Large language models (LLMs) can produce long, coherent passages of text, suggesting that LLMs, although trained on next-word prediction, must represent the latent structure that characterizes a document.
Reconciling Shared versus Context-Specific Information in a Neural Network Model of Latent Causes
Through learning, it naturally stores structure that is shared across tasks in the network weights.
Exploring the hierarchical structure of human plans via program generation
We find that humans are sensitive to both metrics, but that both accounts fail to predict a qualitative feature of human-created programs, namely that people prefer programs with reuse over and above the predictions of MDL.
A Metalearned Neural Circuit for Nonparametric Bayesian Inference
Most applications of machine learning to classification assume a closed set of balanced classes.
Implicit Maximum a Posteriori Filtering via Adaptive Optimization
Bayesian filtering approximates the true underlying behavior of a time-varying system by inverting an explicit generative model to convert noisy measurements into state estimates.
MacGyver: Are Large Language Models Creative Problem Solvers?
We explore the creative problem-solving capabilities of modern LLMs in a novel constrained setting.
Bayes in the age of intelligent machines
The success of methods based on artificial neural networks in creating intelligent machines seems like it might pose a challenge to explanations of human cognition in terms of Bayesian inference.
Improving Interpersonal Communication by Simulating Audiences with Language Models
How do we communicate with others to achieve our goals?
Concept Alignment as a Prerequisite for Value Alignment
Value alignment is essential for building AI systems that can safely and reliably interact with people.
Getting aligned on representational alignment
Finally, we lay out open problems in representational alignment where progress can benefit all three of these fields.
Structurally guided task decomposition in spatial navigation tasks
How are people able to plan so efficiently despite limited cognitive resources?
Dimensions of Disagreement: Unpacking Divergence and Misalignment in Cognitive Science and Artificial Intelligence
The increasing prevalence of artificial agents creates a correspondingly increasing need to manage disagreements between humans and artificial agents, as well as between artificial agents themselves.
Relational Constraints On Neural Networks Reproduce Human Biases towards Abstract Geometric Regularity
Uniquely among primates, humans possess a remarkable capacity to recognize and manipulate abstract structure in the service of task goals across a broad range of behaviors.
Embers of Autoregression: Understanding Large Language Models Through the Problem They are Trained to Solve
This approach - which we call the teleological approach - leads us to identify three factors that we hypothesize will influence LLM accuracy: the probability of the task to be performed, the probability of the target output, and the probability of the provided input.
Cognitive Architectures for Language Agents
Recent efforts have augmented large language models (LLMs) with external resources (e. g., the Internet) or internal control flows (e. g., prompt chaining) for tasks requiring grounding or reasoning, leading to a new class of language agents.
The Universal Law of Generalization Holds for Naturalistic Stimuli
Shepard's universal law of generalization is a remarkable hypothesis about how intelligent organisms should perceive similarity.
Modeling rapid language learning by distilling Bayesian priors into artificial neural networks
We show that learning from limited naturalistic data is possible with an approach that combines the strong inductive biases of a Bayesian model with the flexible representations of a neural network.
Tree of Thoughts: Deliberate Problem Solving with Large Language Models
Language models are increasingly being deployed for general problem solving across a wide range of tasks, but are still confined to token-level, left-to-right decision-making processes during inference.
Neural Constraint Satisfaction: Hierarchical Abstraction for Combinatorial Generalization in Object Rearrangement
Object rearrangement is a challenge for embodied agents because solving these tasks requires generalizing across a combinatorially large set of configurations of entities and their locations.
Superhuman Artificial Intelligence Can Improve Human Decision Making by Increasing Novelty
We then examine human players' strategies across time and find that novel decisions (i. e., previously unobserved moves) occurred more frequently and became associated with higher decision quality after the advent of superhuman AI.
Large language models predict human sensory judgments across six modalities
Determining the extent to which the perceptual world can be recovered from language is a longstanding problem in philosophy and cognitive science.
Humans decompose tasks by trading off utility and computational cost
Human behavior emerges from planning over elaborate decompositions of tasks into goals, subgoals, and low-level actions.
On the Informativeness of Supervision Signals
Supervised learning typically focuses on learning transferable representations from training examples annotated by humans.
Analyzing Diffusion as Serial Reproduction
Diffusion models are a class of generative models that learn to synthesize samples by inverting a diffusion process that gradually maps data into noise.
Bias amplification in experimental social networks is reduced by resampling
Large-scale social networks are thought to contribute to polarization by amplifying people's biases.
Gaussian Process Surrogate Models for Neural Networks
Not being able to understand and predict the behavior of deep learning systems makes it hard to decide what architecture and algorithm to use for a given problem.
Predicting Word Learning in Children from the Performance of Computer Vision Systems
The performance of the computer vision systems is correlated with human judgments of the concreteness of words, which are in turn a predictor of children's word learning, suggesting that these models are capturing the relationship between words and visual phenomena.
Object Representations as Fixed Points: Training Iterative Refinement Algorithms with Implicit Differentiation
Iterative refinement -- start with a random guess, then iteratively improve the guess -- is a useful paradigm for representation learning because it offers a way to break symmetries among equally plausible explanations for the data.
Words are all you need? Language as an approximation for human similarity judgments
Based on the results of this comprehensive study, we provide a concise guide for researchers interested in collecting or approximating human similarity data.
Using Natural Language and Program Abstractions to Instill Human Inductive Biases in Machines
Co-training on these representations result in more human-like behavior in downstream meta-reinforcement learning agents than less abstract controls (synthetic language descriptions, program induction without learned primitives), suggesting that the abstraction supported by these representations is key.
Linguistic communication as (inverse) reward design
We then define a pragmatic listener which performs inverse reward design by jointly inferring the speaker's latent horizon and rewards.
Disentangling Abstraction from Statistical Pattern Matching in Human and Machine Learning
However, because neural networks are hard to interpret, it can be difficult to tell whether agents have learned the underlying abstraction, or alternatively statistical patterns that are characteristic of that abstraction.
Probing BERT's priors with serial reproduction chains
Sampling is a promising bottom-up method for exposing what generative models have learned about language, but it remains unclear how to generate representative samples from popular masked language models (MLMs) like BERT.
Can Humans Do Less-Than-One-Shot Learning?
Being able to learn from small amounts of data is a key characteristic of human intelligence, but exactly {\em how} small?
Predicting Human Similarity Judgments Using Large Language Models
Similarity judgments provide a well-established method for accessing mental representations, with applications in psychology, neuroscience and machine learning.
A pragmatic account of the weak evidence effect
Here, we extend recent probabilistic models of recursive social reasoning to allow for persuasive goals and show that our model provides a pragmatic account for why weakly favorable arguments may backfire, a phenomenon known as the weak evidence effect.
Exploring the Structure of Human Adjective Representations
Human semantic representations are both difficult to capture and hard to fully interpret.
Distinguishing rule- and exemplar-based generalization in learning systems
Machine learning systems often do not share the same inductive biases as humans and, as a result, extrapolate or generalize in ways that are inconsistent with our expectations.
Cognitive science as a source of forward and inverse models of human decisions for robotics and control
Those designing autonomous systems that interact with humans will invariably face questions about how humans think and make decisions.
Passive Attention in Artificial Neural Networks Predicts Human Visual Selectivity
Similarly, we find that recognition performance in the same ANN models was likewise influenced by masking input images using human visual selectivity maps.
Modularity in Reinforcement Learning via Algorithmic Independence in Credit Assignment
Empirical evidence suggests that such action-value methods are more sample efficient than policy-gradient methods on transfer problems that require only sparse changes to a sequence of previously optimal decisions.
Extending rational models of communication from beliefs to actions
Speakers communicate to influence their partner's beliefs and shape their actions.
Are Convolutional Neural Networks or Transformers more like human vision?
Our focus is on comparing a suite of standard Convolutional Neural Networks (CNNs) and a recently-proposed attention-based network, the Vision Transformer (ViT), which relaxes the translation-invariance constraint of CNNs and therefore represents a model with a weaker set of inductive biases.
People construct simplified mental representations to plan
We propose a computational account of this simplification process and, in a series of pre-registered behavioral experiments, show that it is subject to online cognitive control and that people optimally balance the complexity of a task representation and its utility for planning and acting.
Shades of confusion: Lexical uncertainty modulates ad hoc coordination in an interactive communication task
There is substantial variability in the expectations that communication partners bring into interactions, creating the potential for misunderstandings.
From partners to populations: A hierarchical Bayesian account of coordination and convention
Languages are powerful solutions to coordination problems: they provide stable, shared expectations about how the words we say correspond to the beliefs and intentions in our heads.
Evaluating Models of Robust Word Recognition with Serial Reproduction
Spoken communication occurs in a "noisy channel" characterized by high levels of environmental noise, variability within and between speakers, and lexical and syntactic ambiguity.
Show or Tell? Demonstration is More Robust to Changes in Shared Perception than Explanation
Nonetheless, a teacher and learner may not always experience or attend to the same aspects of the environment.
Competition in Cross-situational Word Learning: A Computational Study
Children learn word meanings by tapping into the commonalities across different situations in which words are used and overcome the high level of uncertainty involved in early word learning experiences.
Meta-Learning of Structured Task Distributions in Humans and Machines
We then introduce a novel approach to constructing a "null task distribution" with the same statistical complexity as this structured task distribution but without the explicit rule-based structure used to generate the structured task.
Investigating representations of verb bias in neural language models
Languages typically provide more than one grammatical construction to express certain types of messages.
Learning Rewards from Linguistic Feedback
The sentiment models outperform the inference network, with the "pragmatic" model approaching human performance.
Aspect-Based Sentiment Analysis
Aspect-Based Sentiment Analysis (ABSA)
+2
Understanding Human Intelligence through Human Limitations
Recent progress in artificial intelligence provides the opportunity to ask the question of what is unique about human intelligence, but with a new comparison class.
Resource-rational Task Decomposition to Minimize Planning Costs
That is, rather than planning over a monolithic representation of a task, they decompose the task into simpler subtasks and then plan to accomplish those.
End-to-end Deep Prototype and Exemplar Models for Predicting Human Behavior
The stimulus representations employed in such models are either hand-designed by the experimenter, inferred circuitously from human judgments, or borrowed from pretrained deep neural networks that are themselves competing models of category learning.
Decentralized Reinforcement Learning: Global Decision-Making via Local Economic Transactions
This paper seeks to establish a framework for directing a society of simple, specialized, self-interested agents to solve what traditionally are posed as monolithic single-agent sequential decision problems.
Universal linguistic inductive biases via meta-learning
To facilitate computational modeling aimed at addressing this question, we introduce a framework for giving particular linguistic inductive biases to a neural network model; such a model can then be used to empirically explore the effects of those inductive biases.
Analogy as Nonparametric Bayesian Inference over Relational Systems
Much of human learning and inference can be framed within the computational problem of relational generalization.
Extracting low-dimensional psychological representations from convolutional neural networks
Deep neural networks are increasingly being used in cognitive modeling as a means of deriving representations for complex stimuli such as images.
The Efficiency of Human Cognition Reflects Planned Information Processing
Thus, people should plan their actions, but they should also be smart about how they deploy resources used for planning their actions.
Generalizing meanings from partners to populations: Hierarchical inference supports convention formation on networks
A key property of linguistic conventions is that they hold over an entire community of speakers, allowing us to communicate efficiently even with people we have never met before.
Scaling up Psychology via Scientific Regret Minimization: A Case Study in Moral Decisions
In this paper, we offer a way to enable researchers to systematically build models and identify novel phenomena in large datasets.
On the Utility of Learning about Humans for Human-AI Coordination
While we would like agents that can coordinate with humans, current algorithms such as self-play and population-based training create agents that can coordinate with themselves.
Human uncertainty makes classification more robust
We then show that, while contemporary classifiers fail to exhibit human-like uncertainty on their own, explicit training on our dataset closes this gap, supports improved generalization to increasingly out-of-training-distribution test datasets, and confers robustness to adversarial attacks.
The Computational Structure of Unintentional Meaning
Speech-acts can have literal meaning as well as pragmatic meaning, but these both involve consequences typically intended by a speaker.
Cognitive Model Priors for Predicting Human Decisions
To solve this problem, what is needed are machine learning models with appropriate inductive biases for capturing human behavior, and larger datasets.
Modulating transfer between tasks in gradient-based meta-learning
Learning-to-learn or meta-learning leverages data-driven inductive bias to increase the efficiency of learning on a novel task.
Capturing human categorization of natural images at scale by combining deep networks and cognitive models
Human categorization is one of the most important and successful targets of cognitive modeling in psychology, yet decades of development and assessment of competing models have been contingent on small sets of simple, artificial experimental stimuli.
Predicting human decisions with behavioral theories and machine learning
Here, we introduce BEAST Gradient Boosting (BEAST-GB), a novel hybrid model that synergizes behavioral theories, specifically the model BEAST, with machine learning techniques.
Using Machine Learning to Guide Cognitive Modeling: A Case Study in Moral Reasoning
Large-scale behavioral datasets enable researchers to use complex machine learning algorithms to better predict human behavior, yet this increased predictive power does not always lead to a better understanding of the behavior in question.
Reconciling meta-learning and continual learning with online mixtures of tasks
Learning-to-learn or meta-learning leverages data-driven inductive bias to increase the efficiency of learning on a novel task.
Evaluating Theory of Mind in Question Answering
We propose a new dataset for evaluating question answering models with respect to their capacity to reason about beliefs.
Adaptive Sampling for Convex Regression
In this paper, we introduce the first principled adaptive-sampling procedure for learning a convex function in the $L_\infty$ norm, a problem that arises often in the behavioral and social sciences.
Exploiting Effective Representations for Chinese Sentiment Analysis Using a Multi-Channel Convolutional Neural Network
Effective representation of a text is critical for various natural language processing tasks.
Representational efficiency outweighs action efficiency in human program induction
The importance of hierarchically structured representations for tractable planning has long been acknowledged.
Automatically Composing Representation Transformations as a Means for Generalization
A generally intelligent learner should generalize to more complex tasks than it has previously encountered, but the two common paradigms in machine learning -- either training a separate learner per task or training a single learner for all tasks -- both have difficulty with such generalization because they do not leverage the compositional structure of the task distribution.
Learning a face space for experiments on human identity
Generative models of human identity and appearance have broad applicability to behavioral science and technology, but the exquisite sensitivity of human face perception means that their utility hinges on the alignment of the model's representation to human psychological representations and the photorealism of the generated images.
Learning Hierarchical Visual Representations in Deep Neural Networks Using Hierarchical Linguistic Labels
Modern convolutional neural networks (CNNs) are able to achieve human-level object classification accuracy on specific tasks, and currently outperform competing models in explaining complex human visual representations.
Capturing human category representations by sampling in deep feature spaces
Here, we introduce a method to estimate the structure of human categories that combines ideas from cognitive science and machine learning, blending human-based algorithms with state-of-the-art deep image generators.
Investigating Human Priors for Playing Video Games
What makes humans so good at solving seemingly complex video games?
Generating Plans that Predict Themselves
We introduce $t$-\ACty{}: a measure that quantifies the accuracy and confidence with which human observers can predict the remaining robot plan from the overall task goal and the observed initial $t$ actions in the plan.
Goal Inference Improves Objective and Perceived Performance in Human-Robot Collaboration
The study of human-robot interaction is fundamental to the design and use of robotics in real-world applications.
Learning to select computations
The efficient use of limited computational resources is an essential ingredient of intelligence.
Modeling Human Categorization of Natural Images Using Deep Feature Representations
Over the last few decades, psychologists have developed sophisticated formal models of human categorization using simple artificial stimuli.
Pragmatic-Pedagogic Value Alignment
In robotics, value alignment is key to the design of collaborative robots that can integrate into human workflows, successfully inferring and adapting to their users' objectives as they go.
Evaluating (and improving) the correspondence between deep neural networks and human representations
These networks learn representations of real-world stimuli that can potentially be leveraged to capture psychological representations.
Evaluating vector-space models of analogy
Vector-space representations provide geometric tools for reasoning about the similarity of a set of objects and their relationships.
A rational analysis of curiosity
We present a rational analysis of curiosity, proposing that people's curiosity is driven by seeking stimuli that maximize their ability to make appropriate responses in the future.
Evidence for the size principle in semantic and perceptual domains
Shepard's Universal Law of Generalization offered a compelling case for the first physics-like law in cognitive science that should hold for all intelligent agents in the universe.
Word forms - not just their lengths- are optimized for efficient communication
The inverse relationship between the length of a word and the frequency of its use, first identified by G. K. Zipf in 1935, is a classic empirical law that holds across a wide range of human languages.
Formalizing Neurath's Ship: Approximate Algorithms for Online Causal Learning
Higher-level cognition depends on the ability to learn models of the world.
Adapting Deep Network Features to Capture Psychological Representations
To remedy this, we develop a method for adapting deep features to align with human similarity judgments, resulting in image representations that can potentially be used to extend the scope of psychological experiments.
Infant directed speech is consistent with teaching
We qualitatively compare the simulated teaching data with human IDS, finding that the teaching data exhibit many features of IDS, including some that have been taken as evidence IDS is not for teaching.
Human memory search as a random walk in a semantic network
The human mind has a remarkable ability to store a vast amount of information in memory, and an even more remarkable ability to retrieve these experiences when needed.
Testing a Bayesian Measure of Representativeness Using a Large Image Database
How do people determine which elements of a set are most representative of that set?
An ideal observer model for identifying the reference frame of objects
The object people perceive in an image can depend on its orientation relative to the scene it is in (its reference frame).
A rational model of causal inference with continuous causes
This model successfully predicts human judgments from previous studies better than models of discrete causal inference, and outperforms several other plausible models of causal induction with continuous causes in accounting for people's inferences in a new experiment.
Learning invariant features using the Transformed Indian Buffet Process
Identifying the features of objects becomes a challenge when those features can change in their appearance.
Nonparametric Latent Feature Models for Link Prediction
As the availability and importance of relational data -- such as the friendships summarized on a social networking website -- increases, it becomes increasingly important to have good models for such data.
Neural Implementation of Hierarchical Bayesian Inference by Importance Sampling
Here we propose a simple mechanism for Bayesian inference which involves averaging over a few feature detection neurons which fire at a rate determined by their similarity to a sensory stimulus.
Differential Use of Implicit Negative Evidence in Generative and Discriminative Language Learning
The former analyses assume that learners seek to identify grammatical sentences in a way that is robust to the distribution from which the sentences are generated, analogous to discriminative approaches in machine learning.
A rational model of preference learning and choice prediction by children
Young children demonstrate the ability to make inferences about the preferences of other agents based on their choices.
Modeling human function learning with Gaussian processes
Accounts of how people learn functional relationships between continuous variables have tended to focus on two possibilities: that people are estimating explicit functions, or that they are simply performing associative learning supported by similarity.
Analyzing human feature learning as nonparametric Bayesian inference
Almost all successful machine learning algorithms and cognitive models require powerful representations capturing the features that are relevant to a particular problem.
How memory biases affect information transmission: A rational analysis of serial reproduction
Many human interactions involve pieces of information being passed from one person to another, raising the question of how this process of information transmission is affected by the capacities of the agents involved.
Modeling the effects of memory on human online sentence processing with particle filters
Language comprehension in humans is significantly constrained by memory, yet rapid, highly incremental, and capable of utilizing a wide range of contextual information to resolve ambiguity and form expectations about future input.
A Probabilistic Approach to Language Change
We present a probabilistic approach to language change in which word forms are represented by phoneme sequences that undergo stochastic edits along the branches of a phylogenetic tree.
