1 citations · 1 across the 10 of their papers we have counts for
12 papers
Delay Complexity Dimension: Learning with Delayed and Censored Labels
Madhava Gaikwad
We study PAC learning of binary classifiers when labels arrive after random delays and are censored beyond an observation window . How many samples does this observation model r…
Information Design Against Gaming and Learning Adversaries
Madhava Gaikwad
A principal who deploys a binary classifier with an abstention option must decide which queries the mechanism abstains on. The right choice depends on the adversary. A gaming adver…
A Translational Note on AI Safety Evaluation
Madhava Gaikwad
Recent studies report that automated red-teaming finds more vulnerabilities, at lower cost, than human red-teaming on standard AI safety benchmarks, and some read this as evidence…
The Abstention Protocol: RCA for Clos Fabrics
Madhava Gaikwad, Deepak Pandey
Root cause analysis (RCA) in large datacenter networks is challenging because telemetry is noisy, partial, and asynchronous. Score-based approaches degrade under these conditions,…
Algorithmic Analysis of Dense Associative Memory: Finite-Size Guarantees and Adversarial Robustness
Madhava Gaikwad
Dense Associative Memory (DAM) generalizes Hopfield networks through higher-order interactions and achieves storage capacity that scales as under suitable pattern sepa…
Did You Check the Right Pocket? Cost-Sensitive Store Routing for Memory-Augmented Agents
Madhava Gaikwad
Memory-augmented agents maintain multiple specialized stores, yet most systems retrieve from all stores for every query, increasing cost and introducing irrelevant context. We form…
AlignDP: Hybrid Differential Privacy with Rarity-Aware Protection for LLMs
Madhava Gaikwad
Large language models are exposed to risks of extraction, distillation, and unauthorized fine-tuning. Existing defenses use watermarking or monitoring, but these act after leakage.…
Murphys Laws of AI Alignment: Why the Gap Always Wins
Madhava Gaikwad
We study reinforcement learning from human feedback under misspecification. Sometimes human feedback is systematically wrong on certain types of inputs, like a broken compass that…
AVEC: Bootstrapping Privacy for Local LLMs
Madhava Gaikwad
This position paper presents AVEC (Adaptive Verifiable Edge Control), a framework for bootstrapping privacy for local language models by enforcing privacy at the edge with explicit…
When Are Two RLHF Objectives the Same?
Madhava Gaikwad
The preference optimization literature contains many proposed objectives, often presented as distinct improvements. We introduce Opal, a canonicalization algorithm that determines…
ANSC: Probabilistic Capacity Health Scoring for Datacenter-Scale Reliability
Madhava Gaikwad, Abhishek Gandhi
We present ANSC, a probabilistic capacity health scoring framework for hyperscale datacenter fabrics. While existing alerting systems detect individual device or link failures, the…
NPO: Learning Alignment and Meta-Alignment through Structured Human Feedback
Madhava Gaikwad, Ashwini Ramchandra Doke
We present NPO, an alignment-aware learning framework that operationalizes feedback-driven adaptation in human-in-the-loop decision systems. Unlike prior approaches that treat alig…