arXiv subjects

Alexander Peysakhovich

Publications and source records attributed to Alexander Peysakhovich.

At least 19 recordsLinked to original sources

arXiv2026

Reward Weighted Classifier-Free Guidance as Policy Improvement in Autoregressive Models

Consider an auto-regressive model that produces outputs x (e.g., answers to questions, molecules) each of which can be summarized by an attribute vector y (e.g., helpfulness vs. harmlessness, or bio-availability vs. lipophilicity). An arbitrary reward function r(y) encodes tradeoffs between these properties. Typically, tilting the model's sampling distribution to increase this reward is done at training time via reinforcement learning. However, if the reward function changes, re-alignment requires re-training. In this paper, we show that a reward weighted classifier-free guidance (RCFG) can act as a policy improvement operator in this setting, approximating tilting the sampling distribution by the Q function. We apply RCFG to molecular generation, demonstrating that it can optimize novel reward functions at test time. Finally, we show that using RCFG as a teacher and distilling into the base policy to serve as a warm start significantly speeds up convergence for standard RL.

cs.LG↗

arXiv2025

MONET -- Virtual Cell Painting of Brightfield Images and Time Lapses Using Reference Consistent Diffusion

Cell painting is a popular technique for creating human-interpretable, high-contrast images of cell morphology. There are two major issues with cell paint: (1) it is labor-intensive and (2) it requires chemical fixation, making the study of cell dynamics impossible. We train a diffusion model (Morphological Observation Neural Enhancement Tool, or MONET) on a large dataset to predict cell paint channels from brightfield images. We show that model quality improves with scale. The model uses a consistency architecture to generate time-lapse videos, despite the impossibility of obtaining cell paint video training data. In addition, we show that this architecture enables a form of in-context learning, allowing the model to partially transfer to out-of-distribution cell lines and imaging protocols. Virtual cell painting is not intended to replace physical cell painting completely, but to act as a complementary tool enabling novel workflows in biological research.

cs.CV↗

arXiv2024

MUMU: Bootstrapping Multimodal Image Generation from Text-to-Image Data

We train a model to generate images from multimodal prompts of interleaved text and images such as "a man and his dog in an animated style." We bootstrap a multimodal dataset by extracting semantically meaningful image crops corresponding to words in the image captions of synthetically generated and publicly available text-image data. Our model, MUMU, is composed of a vision-language model encoder with a diffusion decoder and is trained on a single 8xH100 GPU node. Despite being only trained on crops from the same image, MUMU learns to compose inputs from different images into a coherent output. For example, an input of a realistic person and a cartoon will output the same person in the cartoon style, and an input of a standing subject and a scooter will output the subject riding the scooter. As a result, our model generalizes to tasks such as style transfer and character consistency. Our results show the promise of using multimodal models as general purpose controllers for image generation.

cs.CV↗

arXiv2023

Attention Sorting Combats Recency Bias In Long Context Language Models

Current language models often fail to incorporate long contexts efficiently during generation. We show that a major contributor to this issue are attention priors that are likely learned during pre-training: relevant information located earlier in context is attended to less on average. Yet even when models fail to use the information from a relevant document in their response, they still pay preferential attention to that document compared to an irrelevant document at the same position. We leverage this fact to introduce ``attention sorting'': perform one step of decoding, sort documents by the attention they receive (highest attention going last), repeat the process, generate the answer with the newly sorted context. We find that attention sorting improves performance of long context models. Our findings highlight some challenges in using off-the-shelf language models for retrieval augmented generation.

cs.CL↗

arXiv2023

Diagnosis Uncertain Models For Medical Risk Prediction

We consider a patient risk models which has access to patient features such as vital signs, lab values, and prior history but does not have access to a patient's diagnosis. For example, this occurs in a model deployed at intake time for triage purposes. We show that such `all-cause' risk models have good generalization across diagnoses but have a predictable failure mode. When the same lab/vital/history profiles can result from diagnoses with different risk profiles (e.g. E.coli vs. MRSA) the risk estimate is a probability weighted average of these two profiles. This leads to an under-estimation of risk for rare but highly risky diagnoses. We propose a fix for this problem by explicitly modeling the uncertainty in risk prediction coming from uncertainty in patient diagnoses. This gives practitioners an interpretable way to understand patient risk beyond a single risk number.

cs.AI↗

arXiv2023

Pseudo-Euclidean Attract-Repel Embeddings for Undirected Graphs

Dot product embeddings take a graph and construct vectors for nodes such that dot products between two vectors give the strength of the edge. Dot products make a strong transitivity assumption, however, many important forces generating graphs in the real world lead to non-transitive relationships. We remove the transitivity assumption by embedding nodes into a pseudo-Euclidean space - giving each node an attract and a repel vector. The inner product between two nodes is defined by taking the dot product in attract vectors and subtracting the dot product in repel vectors. Pseudo-Euclidean embeddings can compress networks efficiently, allow for multiple notions of nearest neighbors each with their own interpretation, and can be `slotted' into existing models such as exponential family embeddings or graph neural networks for better link prediction.

cs.SI↗

arXiv2023

Implementing Fairness Constraints in Markets Using Taxes and Subsidies

Fisher markets are those where buyers with budgets compete for scarce items, a natural model for many real world markets including online advertising. A market equilibrium is a set of prices and allocations of items such that supply meets demand. We show how market designers can use taxes or subsidies in Fisher markets to ensure that market equilibrium outcomes fall within certain constraints. We show how these taxes and subsidies can be computed even in an online setting where the market designer does not have access to private valuations. We adapt various types of fairness constraints proposed in existing literature to the market case and show who benefits and who loses from these constraints, as well as the extent to which properties of markets including Pareto optimality, envy-freeness, and incentive compatibility are preserved. We find that some prior discussed constraints have few guarantees in terms of who is made better or worse off by their imposition.

cs.GT↗

arXiv2023

Observational Causal Inference in Novel Diseases: A Case Study of COVID-19

A key issue for all observational causal inference is that it relies on an unverifiable assumption - that observed characteristics are sufficient to proxy for treatment confounding. In this paper we argue that in medical cases these conditions are more likely to be met in cases where standardized treatment guidelines do not yet exist. One example of such a situation is the emergence of a novel disease. We study the case of early COVID-19 in New York City hospitals and show that observational analysis of two important thereapeutics, anti-coagulation and steroid therapy, gives results that agree with later guidelines issued via combinations of randomized trials and other evidence. We also argue that observational causal inference cannot be applied mechanically and requires domain expertise by the analyst by showing a cautionary tale of a treatment that appears extremely promising in the data, but the result is due to a quirk of hospital policy.

stat.AP↗

arXiv2022

Backplay: "Man muss immer umkehren"

Model-free reinforcement learning (RL) requires a large number of trials to learn a good policy, especially in environments with sparse rewards. We explore a method to improve the sample efficiency when we have access to demonstrations. Our approach, Backplay, uses a single demonstration to construct a curriculum for a given task. Rather than starting each training episode in the environment's fixed initial state, we start the agent near the end of the demonstration and move the starting point backwards during the course of training until we reach the initial state. Our contributions are that we analytically characterize the types of environments where Backplay can improve training speed, demonstrate the effectiveness of Backplay both in large grid worlds and a complex four player zero-sum game (Pommerman), and show that Backplay compares favorably to other competitive methods known to improve sample efficiency. This includes reward shaping, behavioral cloning, and reverse curriculum generation.

cs.LG↗

arXiv2021

Computing large market equilibria using abstractions

Computing market equilibria is an important practical problem for market design, for example in fair division of items. However, computing equilibria requires large amounts of information (typically the valuation of every buyer for every item) and computing power. We consider ameliorating these issues by applying a method used for solving complex games: constructing a coarsened abstraction of a given market, solving for the equilibrium in the abstraction, and lifting the prices and allocations back to the original market. We show how to bound important quantities such as regret, envy, Nash social welfare, Pareto optimality, and maximin share/proportionality when the abstracted prices and allocations are used in place of the real equilibrium. We then study two abstraction methods of interest for practitioners: (1) filling in unknown valuations using techniques from matrix completion, (2) reducing the problem size by aggregating groups of buyers/items into smaller numbers of representative buyers/items and solving for equilibrium in this coarsened market. We find that in real data allocations/prices that are relatively close to equilibria can be computed from even very coarse abstractions.

cs.GT↗

arXiv2020

Evaluating and Rewarding Teamwork Using Cooperative Game Abstractions

Can we predict how well a team of individuals will perform together? How should individuals be rewarded for their contributions to the team performance? Cooperative game theory gives us a powerful set of tools for answering these questions: the Characteristic Function (CF) and solution concepts like the Shapley Value (SV). There are two major difficulties in applying these techniques to real world problems: first, the CF is rarely given to us and needs to be learned from data. Second, the SV is combinatorial in nature. We introduce a parametric model called cooperative game abstractions (CGAs) for estimating CFs from data. CGAs are easy to learn, readily interpretable, and crucially allow linear-time computation of the SV. We provide identification results and sample complexity bounds for CGA models as well as error bounds in the estimation of the SV using CGAs. We apply our methods to study teams of artificial RL agents as well as real world teams from professional sports.

cs.LG↗

arXiv2020

Discovering Context Effects from Raw Choice Data

Many applications in preference learning assume that decisions come from the maximization of a stable utility function. Yet a large experimental literature shows that individual choices and judgements can be affected by "irrelevant" aspects of the context in which they are made. An important class of such contexts is the composition of the choice set. In this work, our goal is to discover such choice set effects from raw choice data. We introduce an extension of the Multinomial Logit (MNL) model, called the context dependent random utility model (CDM), which allows for a particular class of choice set effects. We show that the CDM can be thought of as a second-order approximation to a general choice system, can be inferred optimally using maximum likelihood and, importantly, is easily interpretable. We apply the CDM to both real and simulated choice data to perform principled exploratory analyses for the presence of choice set effects.

cs.LG↗

arXiv2019

Scalable Fair Division for 'At Most One' Preferences

Allocating multiple scarce items across a set of individuals is an important practical problem. In the case of divisible goods and additive preferences a convex program can be used to find the solution that maximizes Nash welfare (MNW). The MNW solution is equivalent to finding the equilibrium of a market economy (aka. the competitive equilibrium from equal incomes, CEEI) and thus has good properties such as Pareto optimality, envy-freeness, and incentive compatibility in the large. Unfortunately, this equivalence (and nice properties) breaks down for general preference classes. Motivated by real world problems such as course allocation and recommender systems we study the case of additive `at most one' (AMO) preferences - individuals want at most 1 of each item and lotteries are allowed. We show that in this case the MNW solution is still a convex program and importantly is a CEEI solution when the instance gets large but has a `low rank' structure. Thus a polynomial time algorithm can be used to scale CEEI (which is in general PPAD-hard) for AMO preferences. We examine whether the properties guaranteed in the limit hold approximately in finite samples using several real datasets.

cs.GT↗

arXiv2019

Fair Division Without Disparate Impact

We consider the problem of dividing items between individuals in a way that is fair both in the sense of distributional fairness and in the sense of not having disparate impact across protected classes. An important existing mechanism for distributionally fair division is competitive equilibrium from equal incomes (CEEI). Unfortunately, CEEI will not, in general, respect disparate impact constraints. We consider two types of disparate impact measures: requiring that allocations be similar across protected classes and requiring that average utility levels be similar across protected classes. We modify the standard CEEI algorithm in two ways: equitable equilibrium from equal incomes, which removes disparate impact in allocations, and competitive equilibrium from equitable incomes which removes disparate impact in attained utility levels. We show analytically that removing disparate impact in outcomes breaks several of CEEI's desirable properties such as envy, regret, Pareto optimality, and incentive compatibility. By contrast, we can remove disparate impact in attained utility levels without affecting these properties. Finally, we experimentally evaluate the tradeoffs between efficiency, equity, and disparate impact in a recommender-system based market.

cs.GT↗

arXiv2019

Robust Multi-agent Counterfactual Prediction

We consider the problem of using logged data to make predictions about what would happen if we changed the `rules of the game' in a multi-agent system. This task is difficult because in many cases we observe actions individuals take but not their private information or their full reward functions. In addition, agents are strategic, so when the rules change, they will also change their actions. Existing methods (e.g. structural estimation, inverse reinforcement learning) make counterfactual predictions by constructing a model of the game, adding the assumption that agents' behavior comes from optimizing given some goals, and then inverting observed actions to learn agent's underlying utility function (a.k.a. type). Once the agent types are known, making counterfactual predictions amounts to solving for the equilibrium of the counterfactual environment. This approach imposes heavy assumptions such as rationality of the agents being observed, correctness of the analyst's model of the environment/parametric form of the agents' utility functions, and various other conditions to make point identification possible. We propose a method for analyzing the sensitivity of counterfactual conclusions to violations of these assumptions. We refer to this method as robust multi-agent counterfactual prediction (RMAC). We apply our technique to investigating the robustness of counterfactual claims for classic environments in market design: auctions, school choice, and social choice. Importantly, we show RMAC can be used in regimes where point identification is impossible (e.g. those which have multiple equilibria or non-injective maps from type distributions to outcomes).

cs.GT↗

arXiv2019

Learning Existing Social Conventions via Observationally Augmented Self-Play

In order for artificial agents to coordinate effectively with people, they must act consistently with existing conventions (e.g. how to navigate in traffic, which language to speak, or how to coordinate with teammates). A group's conventions can be viewed as a choice of equilibrium in a coordination game. We consider the problem of an agent learning a policy for a coordination game in a simulated environment and then using this policy when it enters an existing group. When there are multiple possible conventions we show that learning a policy via multi-agent reinforcement learning (MARL) is likely to find policies which achieve high payoffs at training time but fail to coordinate with the real group into which the agent enters. We assume access to a small number of samples of behavior from the true convention and show that we can augment the MARL objective to help it find policies consistent with the real group's convention. In three environments from the literature - traffic, communication, and team coordination - we observe that augmenting MARL with a small amount of imitation learning greatly increases the probability that the strategy found by MARL fits well with the existing social convention. We show that this works even in an environment where standard training methods very rarely find the true convention of the agent's partners.

cs.AI↗

arXiv2019

Reinforcement Learning and Inverse Reinforcement Learning with System 1 and System 2

Inferring a person's goal from their behavior is an important problem in applications of AI (e.g. automated assistants, recommender systems). The workhorse model for this task is the rational actor model - this amounts to assuming that people have stable reward functions, discount the future exponentially, and construct optimal plans. Under the rational actor assumption techniques such as inverse reinforcement learning (IRL) can be used to infer a person's goals from their actions. A competing model is the dual-system model. Here decisions are the result of an interplay between a fast, automatic, heuristic-based system 1 and a slower, deliberate, calculating system 2. We generalize the dual system framework to the case of Markov decision problems and show how to compute optimal plans for dual-system agents. We show that dual-system agents exhibit behaviors that are incompatible with rational actor assumption. We show that naive applications of rational-actor IRL to the behavior of dual-system agents can generate wrong inference about the agents' goals and suggest interventions that actually reduce the agent's overall utility. Finally, we adapt a simple IRL algorithm to correctly infer the goals of dual system decision-makers. This allows us to make interventions that help, rather than hinder, the dual-system agent's ability to reach their true goals.

cs.LG↗

arXiv2018

Improving pairwise comparison models using Empirical Bayes shrinkage

Comparison data arises in many important contexts, e.g. shopping, web clicks, or sports competitions. Typically we are given a dataset of comparisons and wish to train a model to make predictions about the outcome of unseen comparisons. In many cases available datasets have relatively few comparisons (e.g. there are only so many NFL games per year) or efficiency is important (e.g. we want to quickly estimate the relative appeal of a product). In such settings it is well known that shrinkage estimators outperform maximum likelihood estimators. A complicating matter is that standard comparison models such as the conditional multinomial logit model are only models of conditional outcomes (who wins) and not of comparisons themselves (who competes). As such, different models of the comparison process lead to different shrinkage estimators. In this work we derive a collection of methods for estimating the pairwise uncertainty of pairwise predictions based on different assumptions about the comparison process. These uncertainty estimates allow us both to examine model uncertainty as well as perform Empirical Bayes shrinkage estimation of the model parameters. We demonstrate that our shrunk estimators outperform standard maximum likelihood methods on real comparison data from online comparison surveys as well as from several sports contexts.

stat.ML↗