arXiv subjects

Raghavendra Addanki

Publications and source records attributed to Raghavendra Addanki.

At least 19 recordsLinked to original sources

arXiv2026

COSAC: Counterfactual Credit Assignment in Sequential Cooperative Teams

In cooperative teams where agents act in a fixed order and share a single team-level reward (multi-agent language systems, sequential robotic tasks), per-agent credit assignment is under-determined. Critic-based approaches scale poorly as the number of agents grows owing to the costly maintenance of joint/factored critic(s), whereas the existing critic-free alternatives have other issues: common credit across agents that couples every agent's signal to teammate noise, importance-sampling corrections for upstream-update staleness that incur variance exponential in team size, or per-agent counterfactual replay that isolates each agent's effect at the price of extra environment or reward calls. We propose COSAC, a critic-free per-agent policy gradient for sequential cooperative teams. COSAC fits an additive per-agent decomposition of the team reward by a single ridge regression on the rollout batch (giving each agent a learning signal decoupled from teammate noise), and computes each agent's counterfactual advantage from fictitious continuations of the current policy (policy forward passes that replace both importance-sampling reweighting and per-agent environment replay, at no extra environment or reward cost). The estimator instantiates the Sequential Aristocrat Utility (SeqAU), our extension of Wolpert and Tumer's (2001) aristocrat utility to sequential teams. We prove bias and variance bounds on SeqAU credits that stay controlled as the team grows. Our controlled study on sequential bandits demonstrates that COSAC attains the lowest advantage MSE and consistently low learning regret across team sizes up to $K = 16$. On the AI2 Reasoning Challenge (ARC) task, where four Qwen3-0.6B agents reason in turn about a grade-school science question, COSAC attains faster convergence than the other critic-free baselines.

cs.LG↗

arXiv2026

GT-SVJ: Generative-Transformer-Based Self-Supervised Video Judge For Efficient Video Reward Modeling

Aligning video generative models with human preferences remains challenging: current approaches rely on Vision-Language Models (VLMs) for reward modeling, but these models struggle to capture subtle temporal dynamics. We propose a fundamentally different approach: repurposing video generative models, which are inherently designed to model temporal structure, as reward models. We present the Generative-Transformer-based Self-Supervised Video Judge (\modelname), a novel evaluation model that transforms state-of-the-art video generation models into powerful temporally-aware reward models. Our key insight is that generative models can be reformulated as energy-based models (EBMs) that assign low energy to high-quality videos and high energy to degraded ones, enabling them to discriminate video quality with remarkable precision when trained via contrastive objectives. To prevent the model from exploiting superficial differences between real and generated videos, we design challenging synthetic negative videos through controlled latent-space perturbations: temporal slicing, feature swapping, and frame shuffling, which simulate realistic but subtle visual degradations. This forces the model to learn meaningful spatiotemporal features rather than trivial artifacts. \modelname achieves state-of-the-art performance on GenAI-Bench and MonteBench using only 30K human-annotations: $6\times$ to $65\times$ fewer than existing VLM-based approaches.

cs.CV↗

arXiv2025

ML-Tool-Bench: Tool-Augmented Planning for ML Tasks

The development of autonomous machine learning (ML) agents capable of end-to-end data science workflows represents a significant frontier in artificial intelligence. These agents must orchestrate complex sequences of data analysis, feature engineering, model selection, and hyperparameter optimization, tasks that require sophisticated planning and iteration. While recent work on building ML agents has explored using large language models (LLMs) for direct code generation, tool-augmented approaches offer greater modularity and reliability. However, existing tool-use benchmarks focus primarily on task-specific tool selection or argument extraction for tool invocation, failing to evaluate the sophisticated planning capabilities required for ML Agents. In this work, we introduce a comprehensive benchmark for evaluating tool-augmented ML agents using a curated set of 61 specialized tools and 15 tabular ML challenges from Kaggle. Our benchmark goes beyond traditional tool-use evaluation by incorporating an in-memory named object management, allowing agents to flexibly name, save, and retrieve intermediate results throughout the workflows. We demonstrate that standard ReAct-style approaches struggle to generate valid tool sequences for complex ML pipelines, and that tree search methods with LLM-based evaluation underperform due to inconsistent state scoring. To address these limitations, we propose two simple approaches: 1) using shaped deterministic rewards with structured textual feedback, and 2) decomposing the original problem into a sequence of sub-tasks, which significantly improves trajectory validity and task performance. Using GPT-4o, our approach improves over ReAct by 16.52 percentile positions, taking the median across all Kaggle challenges. We believe our work provides a foundation for developing more capable tool-augmented planning ML agents.

cs.LG↗

arXiv2025

Leveraging semantic similarity for experimentation with AI-generated treatments

Large Language Models (LLMs) enable a new form of digital experimentation where treatments combine human and model-generated content in increasingly sophisticated ways. The main methodological challenge in this setting is representing these high-dimensional treatments without losing their semantic meaning or rendering analysis intractable. Here, we address this problem by focusing on learning low-dimensional representations that capture the underlying structure of such treatments. These representations enable downstream applications such as guiding generative models to produce meaningful treatment variants and facilitating adaptive assignment in online experiments. We propose double kernel representation learning, which models the causal effect through the inner product of kernel-based representations of treatments and user covariates. We develop an alternating-minimization algorithm that learns these representations efficiently from data and provides convergence guarantees under a low-rank factor model. As an application of this framework, we introduce an adaptive design strategy for online experimentation and demonstrate the method's effectiveness through numerical experiments.

stat.ME↗

arXiv2025

Offline RL by Reward-Weighted Fine-Tuning for Conversation Optimization

Offline reinforcement learning (RL) is a variant of RL where the policy is learned from a previously collected dataset of trajectories and rewards. In our work, we propose a practical approach to offline RL with large language models (LLMs). We recast the problem as reward-weighted fine-tuning, which can be solved using similar techniques to supervised fine-tuning (SFT). To showcase the value of our approach, we apply it to learning short-horizon question-answering policies of a fixed length, where the agent reasons about potential answers or asks clarifying questions. Our work stands in a stark contrast to state-of-the-art methods in this domain, based on SFT and direct preference optimization, which have additional hyper-parameters and do not directly optimize for rewards. We compare to them empirically, and report major gains in both optimized rewards and language quality.

cs.CL↗

arXiv2024

Causal Discovery-Driven Change Point Detection in Time Series

Change point detection in time series aims to identify moments when the probability distribution of time series changes. It is widely applied in many areas, such as human activity sensing and medical science. In the context of multivariate time series, this typically involves examining the joint distribution of multiple variables: If the distribution of any one variable changes, the entire time series undergoes a distribution shift. However, in practical applications, we may be interested only in certain components of the time series, exploring abrupt changes in their distributions while accounting for the presence of other components. Here, assuming an underlying structural causal model that governs the time-series data generation, we address this task by proposing a two-stage non-parametric algorithm that first learns parts of the causal structure through constraint-based discovery methods, and then employs conditional relative Pearson divergence estimation to identify the change points. The conditional relative Pearson divergence quantifies the distribution difference between consecutive segments in the time series, while the causal discovery method allows a focus on the causal mechanism, facilitating access to independent and identically distributed (IID) samples. Theoretically, the typical assumption of samples being IID in conventional change point detection methods can be relaxed based on the Causal Markov Condition. Through experiments on both synthetic and real-world datasets, we validate the correctness and utility of our approach.

cs.LG↗

arXiv2024

Causal Discovery in Semi-Stationary Time Series

Discovering causal relations from observational time series without making the stationary assumption is a significant challenge. In practice, this challenge is common in many areas, such as retail sales, transportation systems, and medical science. Here, we consider this problem for a class of non-stationary time series. The structural causal model (SCM) of this type of time series, called the semi-stationary time series, exhibits that a finite number of different causal mechanisms occur sequentially and periodically across time. This model holds considerable practical utility because it can represent periodicity, including common occurrences such as seasonality and diurnal variation. We propose a constraint-based, non-parametric algorithm for discovering causal relations in this setting. The resulting algorithm, PCMCI$_{\Omega}$, can capture the alternating and recurring changes in the causal mechanisms and then identify the underlying causal graph with conditional independence (CI) tests. We show that this algorithm is sound in identifying causal relations on discrete time series. We validate the algorithm with extensive experiments on continuous and discrete simulated data. We also apply our algorithm to a real-world climate dataset.

cs.LG↗

arXiv2024

Limits of Approximating the Median Treatment Effect

Average Treatment Effect (ATE) estimation is a well-studied problem in causal inference. However, it does not necessarily capture the heterogeneity in the data, and several approaches have been proposed to tackle the issue, including estimating the Quantile Treatment Effects. In the finite population setting containing $n$ individuals, with treatment and control values denoted by the potential outcome vectors $\mathbf{a}, \mathbf{b}$, much of the prior work focused on estimating median$(\mathbf{a}) -$ median$(\mathbf{b})$, where median($\mathbf x$) denotes the median value in the sorted ordering of all the values in vector $\mathbf x$. It is known that estimating the difference of medians is easier than the desired estimand of median$(\mathbf{a-b})$, called the Median Treatment Effect (MTE). The fundamental problem of causal inference -- for every individual $i$, we can only observe one of the potential outcome values, i.e., either the value $a_i$ or $b_i$, but not both, makes estimating MTE particularly challenging. In this work, we argue that MTE is not estimable and detail a novel notion of approximation that relies on the sorted order of the values in $\mathbf{a-b}$. Next, we identify a quantity called variability that exactly captures the complexity of MTE estimation. By drawing connections to instance-optimality studied in theoretical computer science, we show that every algorithm for estimating the MTE obtains an approximation error that is no better than the error of an algorithm that computes variability. Finally, we provide a simple linear time algorithm for computing the variability exactly. Unlike much prior work, a particular highlight of our work is that we make no assumptions about how the potential outcome vectors are generated or how they are correlated, except that the potential outcome values are $k$-ary, i.e., take one of $k$ discrete values.

cs.LG↗

arXiv2024

Continuous Treatment Effects with Surrogate Outcomes

In many real-world causal inference applications, the primary outcomes (labels) are often partially missing, especially if they are expensive or difficult to collect. If the missingness depends on covariates (i.e., missingness is not completely at random), analyses based on fully observed samples alone may be biased. Incorporating surrogates, which are fully observed post-treatment variables related to the primary outcome, can improve estimation in this case. In this paper, we study the role of surrogates in estimating continuous treatment effects and propose a doubly robust method to efficiently incorporate surrogates in the analysis, which uses both labeled and unlabeled data and does not suffer from the above selection bias problem. Importantly, we establish the asymptotic normality of the proposed estimator and show possible improvements on the variance compared with methods that solely use labeled data. Extensive simulations show our methods enjoy appealing empirical performance.

stat.ML↗

arXiv2023

Anytime-Valid Confidence Sequences in an Enterprise A/B Testing Platform

A/B tests are the gold standard for evaluating digital experiences on the web. However, traditional "fixed-horizon" statistical methods are often incompatible with the needs of modern industry practitioners as they do not permit continuous monitoring of experiments. Frequent evaluation of fixed-horizon tests ("peeking") leads to inflated type-I error and can result in erroneous conclusions. We have released an experimentation service on the Adobe Experience Platform based on anytime-valid confidence sequences, allowing for continuous monitoring of the A/B test and data-dependent stopping. We demonstrate how we adapted and deployed asymptotic confidence sequences in a full featured A/B testing platform, describe how sample size calculations can be performed, and how alternate test statistics like "lift" can be analyzed. On both simulated data and thousands of real experiments, we show the desirable properties of using anytime-valid methods instead of traditional approaches.

stat.AP↗

arXiv2022

Sample Constrained Treatment Effect Estimation

Treatment effect estimation is a fundamental problem in causal inference. We focus on designing efficient randomized controlled trials, to accurately estimate the effect of some treatment on a population of $n$ individuals. In particular, we study sample-constrained treatment effect estimation, where we must select a subset of $s \ll n$ individuals from the population to experiment on. This subset must be further partitioned into treatment and control groups. Algorithms for partitioning the entire population into treatment and control groups, or for choosing a single representative subset, have been well-studied. The key challenge in our setting is jointly choosing a representative subset and a partition for that set. We focus on both individual and average treatment effect estimation, under a linear effects model. We give provably efficient experimental designs and corresponding estimators, by identifying connections to discrepancy minimization and leverage-score-based sampling used in randomized numerical linear algebra. Our theoretical results obtain a smooth transition to known guarantees when $s$ equals the population size. We also empirically demonstrate the performance of our algorithms.

cs.LG↗

arXiv2022

Non-Adaptive Edge Counting and Sampling via Bipartite Independent Set Queries

We study the problem of estimating the number of edges in an $n$-vertex graph, accessed via the Bipartite Independent Set query model introduced by Beame et al. (ITCS '18). In this model, each query returns a Boolean, indicating the existence of at least one edge between two specified sets of nodes. We present a non-adaptive algorithm that returns a $(1\pm \epsilon)$ relative error approximation to the number of edges, with query complexity $\tilde O(\epsilon^{-5}\log^{5} n )$, where $\tilde O(\cdot)$ hides $\textrm{poly}(\log \log n)$ dependencies. This is the first non-adaptive algorithm in this setting achieving $\textrm{poly}(1/\epsilon,\log n)$ query complexity. Prior work requires $\Omega(\log^2 n)$ rounds of adaptivity. We avoid this by taking a fundamentally different approach, inspired by work on single-pass streaming algorithms. Moreover, for constant $\epsilon$, our query complexity significantly improves on the best known adaptive algorithm due to Bhattacharya et al. (STACS '22), which requires $O(\epsilon^{-2} \log^{11} n)$ queries. Building on our edge estimation result, we give the first non-adaptive algorithm for outputting a nearly uniformly sampled edge with query complexity $\tilde O(\epsilon^{-6} \log^{6} n)$, improving on the works of Dell et al. (SODA '20) and Bhattacharya et al. (STACS '22), which require $\Omega(\log^3 n)$ rounds of adaptivity. Finally, as a consequence of our edge sampling algorithm, we obtain a $\tilde O(n\log^ 8 n)$ query algorithm for connectivity, using two rounds of adaptivity. This improves on a three-round algorithm of Assadi et al. (ESA '21) and is tight; there is no non-adaptive algorithm for connectivity making $o(n^2)$ queries.

cs.DS↗

arXiv2022

Improved Approximation and Scalability for Fair Max-Min Diversification

Given an $n$-point metric space $(\mathcal{X},d)$ where each point belongs to one of $m=O(1)$ different categories or groups and a set of integers $k_1, \ldots, k_m$, the fair Max-Min diversification problem is to select $k_i$ points belonging to category $i\in [m]$, such that the minimum pairwise distance between selected points is maximized. The problem was introduced by Moumoulidou et al. [ICDT 2021] and is motivated by the need to down-sample large data sets in various applications so that the derived sample achieves a balance over diversity, i.e., the minimum distance between a pair of selected points, and fairness, i.e., ensuring enough points of each category are included. We prove the following results: 1. We first consider general metric spaces. We present a randomized polynomial time algorithm that returns a factor $2$-approximation to the diversity but only satisfies the fairness constraints in expectation. Building upon this result, we present a $6$-approximation that is guaranteed to satisfy the fairness constraints up to a factor $1-\epsilon$ for any constant $\epsilon$. We also present a linear time algorithm returning an $m+1$ approximation with exact fairness. The best previous result was a $3m-1$ approximation. 2. We then focus on Euclidean metrics. We first show that the problem can be solved exactly in one dimension. For constant dimensions, categories and any constant $\epsilon>0$, we present a $1+\epsilon$ approximation algorithm that runs in $O(nk) + 2^{O(k)}$ time where $k=k_1+\ldots+k_m$. We can improve the running time to $O(nk)+ poly(k)$ at the expense of only picking $(1-\epsilon) k_i$ points from category $i\in [m]$. Finally, we present algorithms suitable to processing massive data sets including single-pass data stream algorithms and composable coresets for the distributed processing.

cs.DS↗

arXiv2021

Collaborative Causal Discovery with Atomic Interventions

We introduce a new Collaborative Causal Discovery problem, through which we model a common scenario in which we have multiple independent entities each with their own causal graph, and the goal is to simultaneously learn all these causal graphs. We study this problem without the causal sufficiency assumption, using Maximal Ancestral Graphs (MAG) to model the causal graphs, and assuming that we have the ability to actively perform independent single vertex (or atomic) interventions on the entities. If the $M$ underlying (unknown) causal graphs of the entities satisfy a natural notion of clustering, we give algorithms that leverage this property and recovers all the causal graphs using roughly logarithmic in $M$ number of atomic interventions per entity. These are significantly fewer than $n$ atomic interventions per entity required to learn each causal graph separately, where $n$ is the number of observable nodes in the causal graph. We complement our results with a lower bound and discuss various extensions of our collaborative setting.

cs.LG↗

arXiv2021

How to Design Robust Algorithms using Noisy Comparison Oracle

Metric based comparison operations such as finding maximum, nearest and farthest neighbor are fundamental to studying various clustering techniques such as $k$-center clustering and agglomerative hierarchical clustering. These techniques crucially rely on accurate estimation of pairwise distance between records. However, computing exact features of the records, and their pairwise distances is often challenging, and sometimes not possible. We circumvent this challenge by leveraging weak supervision in the form of a comparison oracle that compares the relative distance between the queried points such as `Is point u closer to v or w closer to x?'. However, it is possible that some queries are easier to answer than others using a comparison oracle. We capture this by introducing two different noise models called adversarial and probabilistic noise. In this paper, we study various problems that include finding maximum, nearest/farthest neighbor search under these noise models. Building upon the techniques we develop for these comparison operations, we give robust algorithms for k-center clustering and agglomerative hierarchical clustering. We prove that our algorithms achieve good approximation guarantees with a high probability and analyze their query complexity. We evaluate the effectiveness and efficiency of our techniques empirically on various real-world datasets.

cs.DS↗

arXiv2020

Intervention Efficient Algorithms for Approximate Learning of Causal Graphs

We study the problem of learning the causal relationships between a set of observed variables in the presence of latents, while minimizing the cost of interventions on the observed variables. We assume access to an undirected graph $G$ on the observed variables whose edges represent either all direct causal relationships or, less restrictively, a superset of causal relationships (identified, e.g., via conditional independence tests or a domain expert). Our goal is to recover the directions of all causal or ancestral relations in $G$, via a minimum cost set of interventions. It is known that constructing an exact minimum cost intervention set for an arbitrary graph $G$ is NP-hard. We further argue that, conditioned on the hardness of approximate graph coloring, no polynomial time algorithm can achieve an approximation factor better than $Θ(\log n)$, where $n$ is the number of observed variables in $G$. To overcome this limitation, we introduce a bi-criteria approximation goal that lets us recover the directions of all but $εn^2$ edges in $G$, for some specified error parameter $ε> 0$. Under this relaxed goal, we give polynomial time algorithms that achieve intervention cost within a small constant factor of the optimal. Our algorithms combine work on efficient intervention design and the design of low-cost separating set systems, with ideas from the literature on graph property testing.

cs.DS↗

arXiv2020

Efficient Intervention Design for Causal Discovery with Latents

We consider recovering a causal graph in presence of latent variables, where we seek to minimize the cost of interventions used in the recovery process. We consider two intervention cost models: (1) a linear cost model where the cost of an intervention on a subset of variables has a linear form, and (2) an identity cost model where the cost of an intervention is the same, regardless of what variables it is on, i.e., the goal is just to minimize the number of interventions. Under the linear cost model, we give an algorithm to identify the ancestral relations of the underlying causal graph, achieving within a $2$-factor of the optimal intervention cost. This approximation factor can be improved to $1+ε$ for any $ε> 0$ under some mild restrictions. Under the identity cost model, we bound the number of interventions needed to recover the entire causal graph, including the latent variables, using a parameterization of the causal graph through a special type of colliders. In particular, we introduce the notion of $p$-colliders, that are colliders between pair of nodes arising from a specific type of conditioning in the causal graph, and provide an upper bound on the number of interventions as a function of the maximum number of $p$-colliders between any two nodes in the causal graph.

cs.LG↗

arXiv2019

Snomed2Vec: Random Walk and Poincaré Embeddings of a Clinical Knowledge Base for Healthcare Analytics

Representation learning methods that transform encoded data (e.g., diagnosis and drug codes) into continuous vector spaces (i.e., vector embeddings) are critical for the application of deep learning in healthcare. Initial work in this area explored the use of variants of the word2vec algorithm to learn embeddings for medical concepts from electronic health records or medical claims datasets. We propose learning embeddings for medical concepts by using graph-based representation learning methods on SNOMED-CT, a widely popular knowledge graph in the healthcare domain with numerous operational and research applications. Current work presents an empirical analysis of various embedding methods, including the evaluation of their performance on multiple tasks of biomedical relevance (node classification, link prediction, and patient state prediction). Our results show that concept embeddings derived from the SNOMED-CT knowledge graph significantly outperform state-of-the-art embeddings, showing 5-6x improvement in ``concept similarity" and 6-20\% improvement in patient diagnosis.

cs.LG↗