arXiv subjects

Rohit Aggarwal

Publications and source records attributed to Rohit Aggarwal.

8 recordsLinked to original sources

arXiv2025

Cluster-based Adaptive Retrieval: Dynamic Context Selection for RAG Applications

Retrieval-Augmented Generation (RAG) enhances large language models (LLMs) by pulling in external material, document, code, manuals, from vast and ever-growing corpora, to effectively answer user queries. The effectiveness of RAG depends significantly on aligning the number of retrieved documents with query characteristics: narrowly focused queries typically require fewer, highly relevant documents, whereas broader or ambiguous queries benefit from retrieving more extensive supporting information. However, the common static top-k retrieval approach fails to adapt to this variability, resulting in either insufficient context from too few documents or redundant information from too many. Motivated by these challenges, we introduce Cluster-based Adaptive Retrieval (CAR), an algorithm that dynamically determines the optimal number of documents by analyzing the clustering patterns of ordered query-document similarity distances. CAR detects the transition point within similarity distances, where tightly clustered, highly relevant documents shift toward less pertinent candidates, establishing an adaptive cut-off that scales with query complexity. On Coinbase's CDP corpus and the public MultiHop-RAG benchmark, CAR consistently picks the optimal retrieval depth and achieves the highest TES score, outperforming every fixed top-k baseline. In downstream RAG evaluations, CAR cuts LLM token usage by 60%, trims end-to-end latency by 22%, and reduces hallucinations by 10% while fully preserving answer relevance. Since integrating CAR into Coinbase's virtual assistant, we've seen user engagement jump by 200%.

cs.IR↗

arXiv2022

RL4ReAl: Reinforcement Learning for Register Allocation

We aim to automate decades of research and experience in register allocation, leveraging machine learning. We tackle this problem by embedding a multi-agent reinforcement learning algorithm within LLVM, training it with the state of the art techniques. We formalize the constraints that precisely define the problem for a given instruction-set architecture, while ensuring that the generated code preserves semantic correctness. We also develop a gRPC based framework providing a modular and efficient compiler interface for training and inference. Our approach is architecture independent: we show experimental results targeting Intel x86 and ARM AArch64. Our results match or out-perform the heavily tuned, production-grade register allocators of LLVM.

cs.LG↗

arXiv2019

IR2Vec: LLVM IR based Scalable Program Embeddings

We propose IR2Vec, a Concise and Scalable encoding infrastructure to represent programs as a distributed embedding in continuous space. This distributed embedding is obtained by combining representation learning methods with flow information to capture the syntax as well as the semantics of the input programs. As our infrastructure is based on the Intermediate Representation (IR) of the source code, obtained embeddings are both language and machine independent. The entities of the IR are modeled as relationships, and their representations are learned to form a seed embedding vocabulary. Using this infrastructure, we propose two incremental encodings:Symbolic and Flow-Aware. Symbolic encodings are obtained from the seed embedding vocabulary, and Flow-Aware encodings are obtained by augmenting the Symbolic encodings with the flow information. We show the effectiveness of our methodology on two optimization tasks (Heterogeneous device mapping and Thread coarsening). Our way of representing the programs enables us to use non-sequential models resulting in orders of magnitude of faster training time. Both the encodings generated by IR2Vec outperform the existing methods in both the tasks, even while using simple machine learning models. In particular, our results improve or match the state-of-the-art speedup in 11/14 benchmark-suites in the device mapping task across two platforms and 53/68 benchmarks in the Thread coarsening task across four different platforms. When compared to the other methods, our embeddings are more scalable, is non-data-hungry, and has betterOut-Of-Vocabulary (OOV) characteristics.

cs.PL↗

arXiv2012

On the Design of Large Scale Wireless Systems (with detailed proofs)

In this paper, we consider the downlink of large OFDMA-based networks and study their performance bounds as a function of the number of - transmitters $B$, users $K$, and resource-blocks $N$. Here, a resource block is a collection of subcarriers such that all such collections, that are disjoint have associated independently fading channels. In particular, we analyze the expected achievable sum-rate as a function of above variables and derive novel upper and lower bounds for a general spatial geometry of transmitters, a truncated path-loss model, and a variety of fading models. We establish the associated scaling laws for dense and extended networks, and propose design guidelines for the regulators to guarantee various QoS constraints and, at the same time, maximize revenue for the service providers. Thereafter, we develop a distributed resource allocation scheme that achieves the same sum-rate scaling as that of the proposed upper bound for a wide range of $K, B, N$. Based on it, we compare low-powered peer-to-peer networks to high-powered single-transmitter networks and give an additional design principle. Finally, we also show how our results can be extended to the scenario where each of the $B$ transmitters have $M (>1)$ co-located antennas.

cs.IT↗

arXiv2012

Performance Bounds and Associated Design Principles for Multi-Cellular Wireless OFDMA Systems (with Detailed Proofs)

In this paper, we consider the downlink of large-scale multi-cellular OFDMA-based networks and study performance bounds of the system as a function of the number of users $K$, the number of base-stations $B$, and the number of resource-blocks $N$. Here, a resource block is a collection of subcarriers such that all such collections, that are disjoint have associated independently fading channels. We derive novel upper and lower bounds on the sum-utility for a general spatial geometry of base stations, a truncated path loss model, and a variety of fading models (Rayleigh, Nakagami-$m$, Weibull, and LogNormal). We also establish the associated scaling laws and show that, in the special case of fixed number of resource blocks, a grid-based network of base stations, and Rayleigh-fading channels, the sum information capacity of the system scales as $Θ(B \log\log K/B)$ for extended networks, and as $O(B \log\log K)$ and $Ω(\log \log K)$ for dense networks. Interpreting these results, we develop some design principles for the service providers along with some guidelines for the regulators in order to achieve provisioning of various QoS guarantees for the end users and, at the same time, maximize revenue for the service providers.

cs.IT↗

arXiv2011

Joint Scheduling and Resource Allocation in OFDMA Downlink Systems via ACK/NAK Feedback

In this paper, we consider the problem of joint scheduling and resource allocation in the OFDMA downlink, with the goal of maximizing an expected long-term goodput-based utility subject to an instantaneous sum-power constraint, and where the feedback to the base station consists only of ACK/NAKs from recently scheduled users. We first establish that the optimal solution is a partially observable Markov decision process (POMDP), which is impractical to implement. In response, we propose a greedy approach to joint scheduling and resource allocation that maintains a posterior channel distribution for every user, and has only polynomial complexity. For frequency-selective channels with Markov time-variation, we then outline a recursive method to update the channel posteriors, based on the ACK/NAK feedback, that is made computationally efficient through the use of particle filtering. To gauge the performance of our greedy approach relative to that of the optimal POMDP, we derive a POMDP performance upper-bound. Numerical experiments show that, for slowly fading channels, the performance of our greedy scheme is relatively close to the upper bound, and much better than fixed-power random user scheduling (FP-RUS), despite its relatively low complexity.

cs.IT↗

arXiv2011

Joint Scheduling and Resource Allocation in the OFDMA Downlink: Utility Maximization under Imperfect Channel-State Information

We consider the problem of simultaneous user-scheduling, power-allocation, and rate-selection in an OFDMA downlink, with the goal of maximizing expected sum-utility under a sum-power constraint. In doing so, we consider a family of generic goodput-based utilities that facilitate, e.g., throughput-based pricing, quality-of-service enforcement, and/or the treatment of practical modulation-and-coding schemes (MCS). Since perfect knowledge of channel state information (CSI) may be difficult to maintain at the base-station, especially when the number of users and/or subchannels is large, we consider scheduling and resource allocation under imperfect CSI, where the channel state is described by a generic probability distribution. First, we consider the "continuous" case where multiple users and/or code rates can time-share a single OFDMA subchannel and time slot. This yields a non-convex optimization problem that we convert into a convex optimization problem and solve exactly using a dual optimization approach. Second, we consider the "discrete" case where only a single user and code rate is allowed per OFDMA subchannel per time slot. For the mixed-integer optimization problem that arises, we discuss the connections it has with the continuous case and show that it can solved exactly in some situations. For the other situations, we present a bound on the optimality gap. For both cases, we provide algorithmic implementations of the obtained solution. Finally, we study, numerically, the performance of the proposed algorithms under various degrees of CSI uncertainty, utilities, and OFDMA system configurations. In addition, we demonstrate advantages relative to existing state-of-the-art algorithms.

cs.IT↗

arXiv2009

Rate Adaptation via Link-Layer Feedback for Goodput Maximization over a Time-Varying Channel

We consider adapting the transmission rate to maximize the goodput, i.e., the amount of data transmitted without error, over a continuous Markov flat-fading wireless channel. In particular, we consider schemes in which transmitter channel state is inferred from degraded causal error-rate feedback, such as packet-level ACK/NAKs in an automatic repeat request (ARQ) system. In such schemes, the choice of transmission rate affects not only the subsequent goodput but also the subsequent feedback, implying that the optimal rate schedule is given by a partially observable Markov decision process (POMDP). Because solution of the POMDP is computationally impractical, we consider simple suboptimal greedy rate assignment and show that the optimal scheme would itself be greedy if the error-rate feedback was non-degraded. Furthermore, we show that greedy rate assignment using non-degraded feedback yields a total goodput that upper bounds that of optimal rate assignment using degraded feedback. We then detail the implementation of the greedy scheme and propose a reduced-complexity greedy scheme that adapts the transmission rate only once per block of packets. We also investigate the performance of the schemes numerically, and show that the proposed greedy scheme achieves steady-state goodputs that are reasonably close to the upper bound on goodput calculated using non-degraded feedback. A similar improvement is obtained in steady-state goodput, drop rate, and average buffer occupancy in the presence of data buffers. We also investigate an upper bound on the performance of optimal rate assignment for a discrete approximation of the channel and show that such quantization leads to a significant loss in achievable goodput.

cs.IT↗