https://ajishpradeep.com/ 2026-10-12 https://ajishpradeep.com/about/ 2026-10-12 https://ajishpradeep.com/cv/ 2026-10-12 https://ajishpradeep.com/blog/ 2026-10-12 https://ajishpradeep.com/work/planogram-vision-ai/ 2026-10-12 https://ajishpradeep.com/work/inpainting-thesis/ 2026-10-12 https://ajishpradeep.com/blog/taking-a-gradient-spike-apart-by-loss-term-and-by-module/ 2026-10-07 https://ajishpradeep.com/blog/the-direction-between-two-close-noisy-points-is-mostly-noise/ 2026-10-07 https://ajishpradeep.com/blog/the-same-weights-two-execution-paths-two-answers/ 2026-10-07 https://ajishpradeep.com/blog/unlabelled-is-not-negative-and-the-usual-metric-hides-it/ 2026-10-07 https://ajishpradeep.com/blog/restarting-adam-moves-every-weight-by-the-full-learning-rate/ 2026-10-07 https://ajishpradeep.com/blog/sampling-with-replacement-sees-about-63-percent-of-your-data/ 2026-10-07 https://ajishpradeep.com/blog/score-perfect-inputs-before-scoring-a-model/ 2026-10-07 https://ajishpradeep.com/blog/softmax-over-one-token-has-zero-gradient-and-adam-moves-the-weights-anyway/ 2026-10-07 https://ajishpradeep.com/blog/derivative-terms-in-a-loss-amplify-noise-by-one-over-h-squared/ 2026-10-07 https://ajishpradeep.com/blog/every-item-call-makes-the-gpu-wait-for-python/ 2026-10-07 https://ajishpradeep.com/blog/freezing-a-layer-does-not-freeze-its-batch-norm/ 2026-10-07 https://ajishpradeep.com/blog/gradient-clipping-does-different-jobs-under-sgd-and-adam/ 2026-10-07 https://ajishpradeep.com/blog/how-far-a-pretrained-backbone-moves-decides-what-it-keeps/ 2026-10-07 https://ajishpradeep.com/blog/adding-an-output-to-a-trained-network-without-moving-the-old-ones/ 2026-10-07 https://ajishpradeep.com/blog/average-precision-scores-thresholds-you-will-never-use/ 2026-10-07 https://ajishpradeep.com/blog/batch-norm-running-statistics-describe-a-network-that-has-moved/ 2026-10-07 https://ajishpradeep.com/blog/cleaning-noisy-labels-with-a-local-median/ 2026-10-07 https://ajishpradeep.com/blog/correlation-is-not-agreement/ 2026-10-07 https://ajishpradeep.com/blog/a-model-that-learns-its-own-uncertainty-chooses-what-to-fit/ 2026-10-07 https://ajishpradeep.com/blog/a-pseudo-labelled-test-set-measures-agreement-with-the-labeller/ 2026-10-07 https://ajishpradeep.com/blog/a-shape-term-cannot-constrain-what-it-is-invariant-to/ 2026-10-07 https://ajishpradeep.com/blog/a-student-gains-from-a-teacher-only-when-the-teacher-knows-more/ 2026-10-07 https://ajishpradeep.com/blog/a-zero-initialised-layer-learns-only-through-a-nonzero-path/ 2026-10-07 https://ajishpradeep.com/blog/a-confidence-head-learns-the-target-you-give-it/ 2026-10-07 https://ajishpradeep.com/blog/a-dense-predictor-cannot-place-what-its-cell-cannot-see/ 2026-10-07 https://ajishpradeep.com/blog/a-hundred-thousand-correlated-samples-can-count-as-five-hundred/ 2026-10-07 https://ajishpradeep.com/blog/a-loss-cannot-fix-what-the-training-data-never-shows/ 2026-10-07 https://ajishpradeep.com/blog/a-checkpoint-is-more-than-its-state-dict/ 2026-10-07