Akshay Mittal

Akshay Mittal, Elyson De La Cruz

Autonomous AI agent ecosystems require stronger mechanisms for secure discovery, identity verification, capability attestation, and policy governance. Current deployments frequently lack (1) uniform agent discovery, (2) cryptographic agent authentication, (3) capability proofs that protect secrets, and (4) enforceable policy controls. This paper presents an implementation-oriented proof of concept for the Agent Name Service (ANS), a DNS-inspired trust layer for AI agent discovery and interoperability in Kubernetes, grounded in the ANS protocol specification~\cite{huang2025ans}. The implementation uses Decentralized Identifiers (DIDs), Verifiable Credentials (VCs), policy-as-code enforcement with Open Policy Agent (OPA), and Kubernetes-native integration patterns (CRDs, admission controls, service mesh integration). In a demo research environment (3-node cluster, 50-agent workflow simulation), we observe sub-10ms response in demonstrated service paths and full success for scripted demo deployment scenarios. We explicitly scope these findings as proof-of-concept evidence rather than production certification. We further provide a threat model, assumptions, and limitations to separate implemented evidence from protocol-defined and roadmap capabilities. The result is an evidence-grounded pathway from ANS protocol concepts to reproducible engineering practice for secure multi-agent systems.

Akshay Mittal, Vinay Venkatesh, Krishna Kandi et al.

The effectiveness of single-model sequential recommendation architectures, while scalable, is often limited when catering to "power users" in sparse or niche domains. Our previous research, PinnerFormerLite, addressed this by using a fixed weighted loss to prioritize specific domains. However, this approach can be sub-optimal, as a single, uniform weight may not be sufficient for domains with very few interactions, where the training signal is easily diluted by the vast, generic dataset. This paper proposes a novel, data-driven approach: a Dynamic Weighted Loss function with comprehensive theoretical foundations and extensive empirical validation. We introduce an adaptive algorithm that adjusts the loss weight for each domain based on its sparsity in the training data, assigning a higher weight to sparser domains and a lower weight to denser ones. This ensures that even rare user interests contribute a meaningful gradient signal, preventing them from being overshadowed. We provide rigorous theoretical analysis including convergence proofs, complexity analysis, and bounds analysis to establish the stability and efficiency of our approach. Our comprehensive empirical validation across four diverse datasets (MovieLens, Amazon Electronics, Yelp Business, LastFM Music) with state-of-the-art baselines (SIGMA, CALRec, SparseEnNet) demonstrates that this dynamic weighting system significantly outperforms all comparison methods, particularly for sparse domains, achieving substantial lifts in key metrics like Recall at 10 and NDCG at 10 while maintaining performance on denser domains and introducing minimal computational overhead.

Akshay Mittal, Gianluca Iaccarino

Coupled partial differential equation (PDE) systems, which often represent multi-physics models, are naturally suited for modular numerical solution methods. However, several challenges yet remain in extending the benefits of modularization practices to the task of uncertainty propagation. Since the cost of each deterministic PDE solve can be usually expected to be quite significant, statistical sampling based methods like Monte-Carlo (MC) are inefficient because they do not take advantage of the mathematical structure of the problem, and suffer for poor convergence properties. On the other hand, even if each module contains a moderate number of uncertain parameters, implementing spectral methods on the combined high-dimensional parameter space can be prohibitively expensive due to the curse of dimensionality. In this work, we present a module-based and efficient intrusive spectral projection (ISP) method for uncertainty propagation. In our proposed method, each subproblem is separated and modularized via block Gauss-Seidel (BGS) techniques, such that each module only needs to tackle the local stochastic parameter space. Moreover, the computational costs are significantly mitigated by constructing reduced chaos approximations of the input data that enter each module. We demonstrate implementations of our proposed method and its computational gains over the standard ISP method using numerical examples.

Akshay Mittal, Gianluca Iaccarino

Multi-physics models governed by coupled partial differential equation (PDE) systems, are naturally suited for partitioned, or modular numerical solution strategies. Although widely used in tackling deterministic coupled models, several challenges arise in extending the benefits of modularization to uncertainty propagation. On one hand, Monte-Carlo (MC) based methods are prohibitively expensive as the cost of each deterministic PDE solve is usually quite large, while on the other hand, even if each module contains a moderate number of uncertain parameters, implementing spectral methods on the combined high-dimensional parameter space can be prohibitively expensive. In this work, we present a reduced non-intrusive spectral projection (NISP) based uncertainty propagation method which separates and modularizes the uncertainty propagation task in each subproblem using block Gauss-Seidel (BGS) techniques. The overall computational costs in the proposed method are also mitigated by constructing reduced approximations of the input data entering each module. These reduced approximations and the corresponding quadrature rules are constructed via simple linear algebra transformations. We describe these components of the proposed algorithm assuming a generalized polynomial chaos (gPC) model of the stochastic solutions. We demonstrate our proposed method and its computational gains over the standard NISP method using numerical examples.