arXiv subjects

Aditya Chatterjee

Publications and source records attributed to Aditya Chatterjee.

3 recordsLinked to original sources

arXiv2026

From Roofline to Ruggedness: Decomposing and Smoothing the GEMM Performance Landscape

Adjacent GEMM problems that differ by a single 128-element step in N can show 30% different throughput. This pervasive performance ruggedness - invisible to roofline analysis and peak-FLOPs intuition, yet dominant for every non-peak workload - is the subject of this paper. We propose performance ruggedness analysis, an analytical framework complementary to roofline: rather than summarizing a GPU with a scalar bound, it treats the full multidimensional performance surface as the object of study, decomposes its texture into mechanism-attributable components, and separates software-removable from hardware-bound losses. The framing is analogous to deep-learning loss landscapes: a continuous quantity (idealized time 2MNK/peak) made rugged by discrete hardware substrates (tiles, sub-groups, cache lines, DRAM channels). We instantiate it on BF16 NN GEMM on Intel Battlemage (Arc B580, sycl-tla) via a 32,768-configuration sweep over (M,N,K) in {128,...,4096}^3. We introduce roughness, the mean absolute step-to-step throughput change, which starts at 16.8 TFLOPs/128-step against an ideal of 2.0. A two-stage stack - best-of-six dynamic tile selection and a novel dynamic-programming padding-and-splitting optimizer (precomputed once, O(1) at runtime) - cuts roughness by 70% and raises mean throughput by 30%. Cross-tile experiments show the residual sawtooth period scales exactly with the tile size, ruling out cache conflicts and attributing the rest to four hardware-bound sources. Finally, we derive the optimal achievable landscape from first principles - datasheet integers alone, no kernel run or simulator - and turn it into an optimality scale (Kernel Optimality Levels) grading any kernel by how much of that landscape it attains and how close its roughness lies to the hardware floor; the production kernel and our optimized stack rate L0 and L2 despite both reporting ~95% of peak.

cs.PF↗

arXiv2026

Event-Grounded Sparse Autoencoders for Vision-Language-Action Policies

Vision-Language-Action (VLA) policies translate language and visual inputs into robot actions, where their hidden representations directly shape closed-loop behavior. However, mechanistic interpretability tools from language and vision-language models do not transfer cleanly to VLAs: outputs are robot actions rather than human-readable tokens, and interventions can only be tested via expensive closed-loop rollouts. We propose an event-grounded interpretability pipeline that anchors SAE feature analysis to behavioral events rather than text contexts. End-effector keyframes are clustered within each task using visual, state, and temporal cues, linking SAE features to behaviorally salient events and, via optional VLM annotations, to semantic context. To our knowledge, our pipeline is among the first to ground SAE-based VLA analysis in closed-loop behavioral events. Across two simulation architectures and a real-robot study, event-grounded ranking yields the strongest causal effects on OpenVLA and transfers to the continuous action chunks of $\pi_{0.5}$. SAE is a sparse but imperfect intervention basis: usability varies with architecture and intervention site, and aggressive intervention reveals safety and interpretability limits. Overall, event-grounded SAE analysis emerges as a practical starting point for behavior-anchored VLA interpretability, motivating future work on SAE features beyond action-aligned coordinates, finer-grained closed-loop evaluation, and safe interventions for high-stakes VLA deployments. Code is available at \url{https://github.com/xc-j/Event-SAE}.

cs.RO↗

arXiv2020

Bias corrected estimators for proportion of true null hypotheses under exponential model: Application of adaptive FDR-controlling in segmented failure data

Two recently introduced model based bias corrected estimators for proportion of true null hypotheses ($\pi_0$) under multiple hypotheses testing scenario have been restructured for exponentially distributed random observations available for each of the common hypotheses. Based on stochastic ordering, a new motivation behind formulation of some related estimators for $\pi_0$ is given. The reduction of bias for the model based estimators are theoretically justified and algorithms for computing the estimators are also presented. The estimators are also used to formulate a popular adaptive multiple testing procedure. Extensive numerical study supports superiority of the bias corrected estimators. We also point out the adverse effect of using the model based bias correction method without proper assessment of the underlying distribution. A case-study is done with a synthetic dataset in connection with reliability and warranty studies to demonstrate the applicability of the procedure, under a non-Gaussian set up. The results obtained are in line with the intuition and experience of the subject expert. An intriguing discussion has been attempted to conclude the article that also indicates the future scope of study.

math.ST↗