41554 blogs · [ { "id": "01a0d904-c57b-73f5-82b7-69e60c2c64ac", "title": "inFERENCe - a machine learning blog", "url": "https://www.inference.vc/", "published_at": "2026-09-25T10:31:31+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e610bde006", "title": "Ferenc Huszar", "url": "https://www.inference.vc/author/ferenc-huszar/", "published_at": "2026-03-12T16:42:12+00:00" }, { "id": "01a08787-1c09-7061-b732-a338af0c8a54", "title": "The Future of Software", "url": "https://www.inference.vc/the-future-of-software/", "published_at": "2026-02-25T16:27:41+00:00" }, { "id": "01a08787-1c09-7061-b732-a338af30977c", "title": "Deep Learning is Powerful Because It Makes Hard Things Easy - Reflections 10 Years On", "url": "https://www.inference.vc/deep-learning-is-powerful-because-it-makes-hard-things-easy-reflections-10-years-on/", "published_at": "2026-01-31T15:21:52+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e60cc6d9ee", "title": "Ferenc Huszár", "url": "https://www.inference.vc/about/", "published_at": "2025-11-28T12:38:35+00:00" }, { "id": "01a08787-1c09-7061-b732-a338afba82a2", "title": "Discrete Diffusion: Continuous-Time Markov Chains", "url": "https://www.inference.vc/discrete-diffusion-continuous-time-markov-chains/", "published_at": "2025-05-22T09:12:09+00:00" }, { "id": "01a08787-1c09-7061-b732-a338afbcbe17", "title": "We may finally crack Maths. But should we?", "url": "https://www.inference.vc/we-may-finally-crack-maths-but-should-we/", "published_at": "2023-06-08T15:58:13+00:00" }, { "id": "01a08787-1c09-7061-b732-a338b0ac3384", "title": "Mortal Komputation: On Hinton's argument for superhuman AI.", "url": "https://www.inference.vc/mortal-computation-hintons/", "published_at": "2023-05-30T13:48:52+00:00" }, { "id": "01a0d904-c57c-73ac-a6a0-31e715c88239", "title": null, "url": "https://www.inference.vc/tag/agi/", "published_at": "2023-05-30T13:31:20+00:00" }, { "id": "01a0d904-c57c-73ac-a6a0-31e71614663b", "title": null, "url": "https://www.inference.vc/tag/opinion/", "published_at": "2023-05-30T13:31:20+00:00" }, { "id": "01a08787-1c09-7061-b732-a338b162d097", "title": "Autoregressive Models, OOD Prompts and the Interpolation Regime", "url": "https://www.inference.vc/autoregressive-models-in-out-of-distribution/", "published_at": "2023-03-30T11:54:26+00:00" }, { "id": "01a08787-1c09-7061-b732-a338b1ffa9c5", "title": "We May be Surprised Again: Why I take LLMs seriously.", "url": "https://www.inference.vc/we-may-be-surprised-again/", "published_at": "2023-03-22T14:55:54+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e60cd7b4e8", "title": null, "url": "https://www.inference.vc/mlss/", "published_at": "2023-01-11T10:02:33+00:00" }, { "id": "01a08787-1c09-7061-b732-a338b22c0b65", "title": "Implicit Bayesian Inference in Large Language Models", "url": "https://www.inference.vc/implicit-bayesian-inference-in-sequence-models/", "published_at": "2022-03-03T13:57:26+00:00" }, { "id": "01a08787-1c09-7061-b732-a338b27766d2", "title": "Eastern European Guide to Writing Reference Letters", "url": "https://www.inference.vc/the-east-european-guide-to-writing-reference-letters/", "published_at": "2022-02-28T14:29:15+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e610e7235b", "title": null, "url": "https://www.inference.vc/author/patrik/", "published_at": "2021-06-10T17:00:31+00:00" }, { "id": "01a0d904-c57c-73ac-a6a0-31e716d069ad", "title": null, "url": "https://www.inference.vc/tag/causal-inference/", "published_at": "2021-06-10T14:05:39+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e617feb301", "title": null, "url": "https://www.inference.vc/causal-inference-3-counterfactuals/", "published_at": "2021-06-10T14:02:32+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e618a8f78d", "title": null, "url": "https://www.inference.vc/causal-inference-2-illustrating-interventions-in-a-toy-example/", "published_at": "2021-06-10T14:02:09+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e61984a4ad", "title": null, "url": "https://www.inference.vc/untitled/", "published_at": "2021-06-10T14:01:26+00:00" }, { "id": "01a08787-1c09-7061-b732-a338b2d123d6", "title": "Causal inference 4: Causal Diagrams, Markov Factorization, Structural Equation Models", "url": "https://www.inference.vc/causal-inference-4/", "published_at": "2021-06-10T13:56:20+00:00" }, { "id": "01a08787-1c09-7061-b732-a338b39be54e", "title": "On Information Theoretic Bounds for SGD", "url": "https://www.inference.vc/on-information-theoretic-bounds-for-sgd/", "published_at": "2021-04-23T14:17:49+00:00" }, { "id": "01a0d904-c57c-73ac-a6a0-31e717c67f9e", "title": null, "url": "https://www.inference.vc/tag/information-theory/", "published_at": "2021-04-23T14:17:49+00:00" }, { "id": "01a08787-1c09-7061-b732-a338b3ac05c9", "title": "Notes on the Origin of Implicit Regularization in SGD", "url": "https://www.inference.vc/notes-on-the-origin-of-implicit-regularization-in-stochastic-gradient-descent/", "published_at": "2021-04-01T14:23:24+00:00" }, { "id": "01a0d904-c57c-73ac-a6a0-31e717dda451", "title": null, "url": "https://www.inference.vc/tag/generalization/", "published_at": "2021-04-01T14:23:24+00:00" }, { "id": "01a0d904-c57c-73ac-a6a0-31e71869bd71", "title": null, "url": "https://www.inference.vc/tag/sgd/", "published_at": "2021-04-01T14:23:24+00:00" }, { "id": "01a0d904-c57c-73ac-a6a0-31e718e983ac", "title": null, "url": "https://www.inference.vc/tag/differerntial-equations/", "published_at": "2021-04-01T14:23:24+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e6111d4f12", "title": null, "url": "https://www.inference.vc/author/dora/", "published_at": "2021-03-18T15:31:27+00:00" }, { "id": "01a08787-1c09-7061-b732-a338b458002c", "title": "An information maximization view on the $\\beta$-VAE objective", "url": "https://www.inference.vc/beta-vae/", "published_at": "2021-03-18T15:18:12+00:00" }, { "id": "01a0d904-c57c-73ac-a6a0-31e7192e3e82", "title": null, "url": "https://www.inference.vc/tag/vae/", "published_at": "2021-03-18T15:18:12+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e61b2d0b98", "title": null, "url": "https://www.inference.vc/deepsets-modeling-permutation-invariance/", "published_at": "2021-03-05T15:42:05+00:00" }, { "id": "01a08787-1c09-7061-b732-a338b525a4d5", "title": "Some Intuition on the Neural Tangent Kernel", "url": "https://www.inference.vc/neural-tangent-kernels-some-intuition-for-kernel-gradient-descent/", "published_at": "2020-11-20T13:57:12+00:00" }, { "id": "01a08787-1c09-7061-b732-a338b5280ca6", "title": "Notes on Causally Correct Partial Models", "url": "https://www.inference.vc/notes-on-causally-correct-partial-models-2/", "published_at": "2020-11-12T14:58:37+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e6118e31ae", "title": null, "url": "https://www.inference.vc/author/fabian-fuchs/", "published_at": "2020-09-17T03:26:37+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e60d4bedee", "title": null, "url": "https://www.inference.vc/information-for-prospective-phd-students/", "published_at": "2020-09-10T13:38:04+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e61d0a97a8", "title": null, "url": "https://www.inference.vc/meta-learning-with-the-implicit-function-theorem/", "published_at": "2019-11-14T18:46:30+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e61d4adadc", "title": null, "url": "https://www.inference.vc/the-secular-bayesian-using-belief-distributions-without-really-believing/", "published_at": "2019-10-31T15:20:42+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e61dcba0a2", "title": null, "url": "https://www.inference.vc/exponentially-growing-learning-rate-implications-of-scale-invariance-induced-by-batchnorm/", "published_at": "2019-10-25T13:02:15+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e61e9d1a41", "title": null, "url": "https://www.inference.vc/marginal-likelihood-and-cross-validation/", "published_at": "2019-10-17T14:07:47+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e61f930512", "title": null, "url": "https://www.inference.vc/notes-on-imaml-meta-learning-without-differentiating-through/", "published_at": "2019-09-20T14:14:13+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e62068dbba", "title": null, "url": "https://www.inference.vc/invariant-risk-minimization/", "published_at": "2019-07-24T11:22:15+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e620bf63af", "title": null, "url": "https://www.inference.vc/my-notes-on-the-numerics-of-gans/", "published_at": "2019-07-09T14:43:06+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e620f0db23", "title": null, "url": "https://www.inference.vc/icml-highlight-contrastive-divergence-for-variational-inference-and-mcmc/", "published_at": "2019-06-12T15:16:13+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e62177c271", "title": null, "url": "https://www.inference.vc/blessings-of-multiple-causes-causal-inference-when-you-cant-measure-confounders/", "published_at": "2019-06-12T14:06:45+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e621c28f7e", "title": null, "url": "https://www.inference.vc/on-empirical-fisher-information/", "published_at": "2019-06-07T08:23:05+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e621fd521b", "title": null, "url": "https://www.inference.vc/perceptual-streightening-of-natural-videos/", "published_at": "2019-05-30T13:08:39+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e62255daa7", "title": null, "url": "https://www.inference.vc/online-bayesian-deep-learning-in-production-at-tencent/", "published_at": "2018-11-15T15:23:09+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e6233c4eb2", "title": null, "url": "https://www.inference.vc/halloween-special-critical-reviews-of-my-least-favourite-nips-2018-submissions/", "published_at": "2018-10-30T12:08:34+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e624381120", "title": null, "url": "https://www.inference.vc/high-dimensional-gaussian-distributions-are-soap-bubble/", "published_at": "2018-05-24T15:50:12+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e624d60db1", "title": null, "url": "https://www.inference.vc/the-lottery-ticket-hypothesis/", "published_at": "2018-05-10T13:47:32+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e625c90f67", "title": null, "url": "https://www.inference.vc/goals-and-principles-of-representation-learning/", "published_at": "2018-04-12T13:51:41+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e6263257e7", "title": null, "url": "https://www.inference.vc/pruning-neural-networks-two-recent-papers/", "published_at": "2018-02-06T19:32:56+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e6272c2b30", "title": null, "url": "https://www.inference.vc/generalization-and-the-fisher-rao-norm-2/", "published_at": "2018-01-26T20:28:50+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e627511195", "title": null, "url": "https://www.inference.vc/sharp-vs-flat-minima-are-still-a-mystery-to-me/", "published_at": "2018-01-19T18:11:20+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e62796e62c", "title": null, "url": "https://www.inference.vc/my-thoughts-on-alchemy/", "published_at": "2017-12-07T17:04:19+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e628614439", "title": null, "url": "https://www.inference.vc/grosses-challenge/", "published_at": "2017-11-29T14:51:23+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e629308b41", "title": null, "url": "https://www.inference.vc/gans-are-being-fixed-in-more-than-one-way/", "published_at": "2017-11-23T15:43:53+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e6294c6796", "title": null, "url": "https://www.inference.vc/design-patterns/", "published_at": "2017-11-17T11:19:03+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e629b27333", "title": null, "url": "https://www.inference.vc/mixup-data-dependent-data-augmentation/", "published_at": "2017-11-03T12:36:36+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e62a300983", "title": null, "url": "https://www.inference.vc/alphago-zero-policy-improvement-and-vector-fields/", "published_at": "2017-10-26T14:48:21+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e62b1f438e", "title": null, "url": "https://www.inference.vc/exchangeable-processes-via-neural-networks/", "published_at": "2017-09-04T10:10:03+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e62c1ea417", "title": null, "url": "https://www.inference.vc/from-instance-noise-to-gradient-regularisation/", "published_at": "2017-06-02T10:01:43+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e62cdc6344", "title": null, "url": "https://www.inference.vc/everything-that-works-works-because-its-bayesian-2/", "published_at": "2017-06-01T15:15:26+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e62d890a28", "title": null, "url": "https://www.inference.vc/exemplar-cnns-and-information-maximization/", "published_at": "2017-05-27T21:04:11+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e62e3b2f24", "title": null, "url": "https://www.inference.vc/maximum-likelihood-for-representation-learning-2/", "published_at": "2017-05-04T19:31:52+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e62edd58fd", "title": null, "url": "https://www.inference.vc/unsupervised-learning-by-predicting-noise-an-information-maximization-view-2/", "published_at": "2017-04-24T20:52:55+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e62f27da74", "title": null, "url": "https://www.inference.vc/evolution-strategies-variational-optimisation-and-natural-es-2/", "published_at": "2017-04-06T19:42:05+00:00" }, { "id": "01a0d904-c57c-73ac-a6a0-31e6f86894a0", "title": null, "url": "https://www.inference.vc/evolutionary-strategies-embarrassingly-parallelizable-optimization/", "published_at": "2017-04-05T13:07:23+00:00" }, { "id": "01a0d904-c57c-73ac-a6a0-31e6f90dc5cd", "title": null, "url": "https://www.inference.vc/choice-of-recognition-models-in-vaes-a-regularisation-view/", "published_at": "2017-03-30T08:05:08+00:00" }, { "id": "01a0d904-c57c-73ac-a6a0-31e6f9765306", "title": null, "url": "https://www.inference.vc/comment-on-overcoming-catastrophic-forgetting-in-nns-are-multiple-penalties-needed-2/", "published_at": "2017-03-16T20:50:53+00:00" }, { "id": "01a0d904-c57c-73ac-a6a0-31e6fa340a11", "title": null, "url": "https://www.inference.vc/the-spherical-kernel-divergence/", "published_at": "2017-03-09T16:58:56+00:00" }, { "id": "01a0d904-c57c-73ac-a6a0-31e6faf47748", "title": null, "url": "https://www.inference.vc/variational-inference-using-implicit-models/", "published_at": "2017-03-01T10:30:29+00:00" }, { "id": "01a0d904-c57c-73ac-a6a0-31e6fb876d90", "title": null, "url": "https://www.inference.vc/variational-renyi-lower-bound/", "published_at": "2017-02-11T14:02:15+00:00" }, { "id": "01a0d904-c57c-73ac-a6a0-31e6fbbef305", "title": null, "url": "https://www.inference.vc/variational-inference-using-implicit-models-part-iii-joint-contrastive-inference-ali-and-bigan/", "published_at": "2017-02-09T20:36:58+00:00" }, { "id": "01a0d904-c57c-73ac-a6a0-31e6fc94b6da", "title": null, "url": "https://www.inference.vc/variational-inference-with-implicit-probabilistic-models-part-1-2/", "published_at": "2017-02-09T14:13:43+00:00" }, { "id": "01a0d904-c57c-73ac-a6a0-31e6fcfe5f74", "title": null, "url": "https://www.inference.vc/variational-inference-with-implicit-models-part-ii-amortised-inference-2/", "published_at": "2017-02-09T14:13:27+00:00" }, { "id": "01a0d904-c57c-73ac-a6a0-31e6fdf0be8a", "title": null, "url": "https://www.inference.vc/variational-inference-using-implicit-models-part-iv-denoisers-instead-of-discriminators/", "published_at": "2017-02-09T14:13:09+00:00" }, { "id": "01a0d904-c57c-73ac-a6a0-31e6fe87f0b7", "title": null, "url": "https://www.inference.vc/holiday-special-deriving-the-subpixel-cnn-from-first-principles/", "published_at": "2016-12-23T16:09:29+00:00" }, { "id": "01a0d904-c57c-73ac-a6a0-31e6fea8c19e", "title": null, "url": "https://www.inference.vc/my-summary-of-adversarial-training-nips-workshop/", "published_at": "2016-12-14T15:36:38+00:00" }, { "id": "01a0d904-c57c-73ac-a6a0-31e6ff18e8ee", "title": null, "url": "https://www.inference.vc/solve-intelligence/", "published_at": "2016-11-24T16:13:33+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e60da14280", "title": null, "url": "https://www.inference.vc/representation-learning-and-compression-with-the-information-bottleneck/", "published_at": "2016-11-11T15:48:20+00:00" }, { "id": "01a0d904-c57c-73ac-a6a0-31e6fff3c4cd", "title": null, "url": "https://www.inference.vc/how-to-train-your-generative-models-why-generative-adversarial-networks-work-so-well-2/", "published_at": "2016-10-24T09:47:15+00:00" }, { "id": "01a0d904-c57c-73ac-a6a0-31e6fff7ca5b", "title": null, "url": "https://www.inference.vc/instance-noise-a-trick-for-stabilising-gan-training/", "published_at": "2016-10-21T10:24:28+00:00" }, { "id": "01a0d904-c57c-73ac-a6a0-31e70030b5ad", "title": null, "url": "https://www.inference.vc/how-powerful-are-graph-convolutions-review-of-kipf-welling-2016-2/", "published_at": "2016-10-03T09:29:25+00:00" }, { "id": "01a0d904-c57c-73ac-a6a0-31e70097523a", "title": null, "url": "https://www.inference.vc/are-energy-based-gans-actually-energy-based/", "published_at": "2016-09-23T10:12:19+00:00" }, { "id": "01a0d904-c57c-73ac-a6a0-31e700c06e93", "title": null, "url": "https://www.inference.vc/infogan-variational-bound-on-mutual-information-twice/", "published_at": "2016-08-09T14:14:57+00:00" }, { "id": "01a0d904-c57c-73ac-a6a0-31e7018a1e34", "title": null, "url": "https://www.inference.vc/temporal-contrastive-learning-for-latent-variable-models/", "published_at": "2016-07-14T14:42:51+00:00" }, { "id": "01a0d904-c57c-73ac-a6a0-31e701bfea11", "title": null, "url": "https://www.inference.vc/understanding-minibatch-discrimination-in-gans/", "published_at": "2016-06-15T14:52:56+00:00" }, { "id": "01a0d904-c57c-73ac-a6a0-31e7026c1dfd", "title": null, "url": "https://www.inference.vc/dilated-convolutions-and-kronecker-factorisation/", "published_at": "2016-05-12T14:59:29+00:00" }, { "id": "01a0d904-c57c-73ac-a6a0-31e702e2dbf1", "title": null, "url": "https://www.inference.vc/notes-on-unsupervised-learning-of-visual-representations-by-solving-jigsaw-puzzles/", "published_at": "2016-04-28T14:04:45+00:00" }, { "id": "01a0d904-c57c-73ac-a6a0-31e7033ae02e", "title": null, "url": "https://www.inference.vc/adversarial-preference-loss/", "published_at": "2016-03-25T14:31:57+00:00" }, { "id": "01a0d904-c57c-73ac-a6a0-31e719ad2ad2", "title": null, "url": "https://www.inference.vc/tag/preference-learning/", "published_at": "2016-03-25T14:31:04+00:00" }, { "id": "01a0d904-c57c-73ac-a6a0-31e7038d03a9", "title": null, "url": "https://www.inference.vc/an-alternative-update-rule-for-generative-adversarial-networks/", "published_at": "2016-03-24T10:37:06+00:00" }, { "id": "01a0d904-c57c-73ac-a6a0-31e71a6be7d0", "title": null, "url": "https://www.inference.vc/tag/generative-models/", "published_at": "2016-03-24T10:36:38+00:00" }, { "id": "01a0d904-c57c-73ac-a6a0-31e71b59757c", "title": null, "url": "https://www.inference.vc/tag/adversarial/", "published_at": "2016-03-24T10:36:38+00:00" }, { "id": "01a0d904-c57c-73ac-a6a0-31e71b8fc9a4", "title": null, "url": "https://www.inference.vc/tag/gan/", "published_at": "2016-03-24T10:36:38+00:00" }, { "id": "01a0d904-c57c-73ac-a6a0-31e71c08f2d1", "title": null, "url": "https://www.inference.vc/tag/kl-divergence/", "published_at": "2016-03-24T10:36:38+00:00" }, { "id": "01a0d904-c57c-73ac-a6a0-31e7040ca92f", "title": null, "url": "https://www.inference.vc/deep-learning-is-easy/", "published_at": "2016-02-01T17:20:23+00:00" }, { "id": "01a0d904-c57b-73f5-82b7-69e60e89847b", "title": null, "url": "https://www.inference.vc/denoising-as-unsupervised-learning/", "published_at": "2016-01-21T16:57:23+00:00" }, { "id": "01a0d904-c57c-73ac-a6a0-31e7042a3c4d", "title": null, "url": "https://www.inference.vc/the-next-episode-2/", "published_at": "2016-01-08T16:37:05+00:00" } ] posts Claim your blog

Back to inference.vc

Blog · corpus.blog/blogs/inference.vc/posts

inference.vc

inference.vc

2026

inFERENCe - a machine learning blog original ↗

25 Sept 2026

Ferenc Huszar original ↗

12 Mar 2026

The Future of Software original ↗

25 Feb 2026

Deep Learning is Powerful Because It Makes Hard Things Easy - Reflections 10 Years On original ↗

31 Jan 2026

2025

Ferenc Huszár original ↗

28 Nov 2025

Discrete Diffusion: Continuous-Time Markov Chains original ↗

22 May 2025

2023

We may finally crack Maths. But should we?original ↗

8 Jun 2023

Mortal Komputation: On Hinton's argument for superhuman AI.original ↗

30 May 2023

https://www.inference.vc/tag/agi/original ↗

30 May 2023

https://www.inference.vc/tag/opinion/original ↗

30 May 2023

Autoregressive Models, OOD Prompts and the Interpolation Regime original ↗

30 Mar 2023

We May be Surprised Again: Why I take LLMs seriously.original ↗

22 Mar 2023

https://www.inference.vc/mlss/original ↗

11 Jan 2023

2022

Implicit Bayesian Inference in Large Language Models original ↗

3 Mar 2022

Eastern European Guide to Writing Reference Letters original ↗

28 Feb 2022

2021

https://www.inference.vc/author/patrik/original ↗

10 Jun 2021

https://www.inference.vc/tag/causal-inference/original ↗

10 Jun 2021

https://www.inference.vc/causal-inference-3-counterfactuals/original ↗

10 Jun 2021

https://www.inference.vc/causal-inference-2-illustrating-interventions-in-a-toy-example/original ↗

10 Jun 2021

https://www.inference.vc/untitled/original ↗

10 Jun 2021

Causal inference 4: Causal Diagrams, Markov Factorization, Structural Equation Models original ↗

10 Jun 2021

On Information Theoretic Bounds for SGD original ↗

23 Apr 2021

https://www.inference.vc/tag/information-theory/original ↗

23 Apr 2021

Notes on the Origin of Implicit Regularization in SGD original ↗

1 Apr 2021

https://www.inference.vc/tag/generalization/original ↗

1 Apr 2021

https://www.inference.vc/tag/sgd/original ↗

1 Apr 2021

https://www.inference.vc/tag/differerntial-equations/original ↗

1 Apr 2021

https://www.inference.vc/author/dora/original ↗

18 Mar 2021

An information maximization view on the $\beta$-VAE objective original ↗

18 Mar 2021

https://www.inference.vc/tag/vae/original ↗

18 Mar 2021

https://www.inference.vc/deepsets-modeling-permutation-invariance/original ↗

5 Mar 2021

2020

Some Intuition on the Neural Tangent Kernel original ↗

20 Nov 2020

Notes on Causally Correct Partial Models original ↗

12 Nov 2020

https://www.inference.vc/author/fabian-fuchs/original ↗

17 Sept 2020

https://www.inference.vc/information-for-prospective-phd-students/original ↗

10 Sept 2020

2019

https://www.inference.vc/meta-learning-with-the-implicit-function-theorem/original ↗

14 Nov 2019

https://www.inference.vc/the-secular-bayesian-using-belief-distributions-without-really-believing/original ↗

31 Oct 2019

https://www.inference.vc/exponentially-growing-learning-rate-implications-of-scale-invariance-induced-by-batchnorm/original ↗

25 Oct 2019

https://www.inference.vc/marginal-likelihood-and-cross-validation/original ↗

17 Oct 2019

https://www.inference.vc/notes-on-imaml-meta-learning-without-differentiating-through/original ↗

20 Sept 2019

https://www.inference.vc/invariant-risk-minimization/original ↗

24 Jul 2019

https://www.inference.vc/my-notes-on-the-numerics-of-gans/original ↗

9 Jul 2019

https://www.inference.vc/icml-highlight-contrastive-divergence-for-variational-inference-and-mcmc/original ↗

12 Jun 2019

https://www.inference.vc/blessings-of-multiple-causes-causal-inference-when-you-cant-measure-confounders/original ↗

12 Jun 2019

https://www.inference.vc/on-empirical-fisher-information/original ↗

7 Jun 2019

https://www.inference.vc/perceptual-streightening-of-natural-videos/original ↗

30 May 2019

2018

https://www.inference.vc/online-bayesian-deep-learning-in-production-at-tencent/original ↗

15 Nov 2018

https://www.inference.vc/halloween-special-critical-reviews-of-my-least-favourite-nips-2018-submissions/original ↗

30 Oct 2018

https://www.inference.vc/high-dimensional-gaussian-distributions-are-soap-bubble/original ↗

24 May 2018

https://www.inference.vc/the-lottery-ticket-hypothesis/original ↗

10 May 2018

https://www.inference.vc/goals-and-principles-of-representation-learning/original ↗

12 Apr 2018

https://www.inference.vc/pruning-neural-networks-two-recent-papers/original ↗

6 Feb 2018

https://www.inference.vc/generalization-and-the-fisher-rao-norm-2/original ↗

26 Jan 2018

https://www.inference.vc/sharp-vs-flat-minima-are-still-a-mystery-to-me/original ↗

19 Jan 2018

2017

https://www.inference.vc/my-thoughts-on-alchemy/original ↗

7 Dec 2017

https://www.inference.vc/grosses-challenge/original ↗

29 Nov 2017

https://www.inference.vc/gans-are-being-fixed-in-more-than-one-way/original ↗

23 Nov 2017

https://www.inference.vc/design-patterns/original ↗

17 Nov 2017

https://www.inference.vc/mixup-data-dependent-data-augmentation/original ↗

3 Nov 2017

https://www.inference.vc/alphago-zero-policy-improvement-and-vector-fields/original ↗

26 Oct 2017

https://www.inference.vc/exchangeable-processes-via-neural-networks/original ↗

4 Sept 2017

https://www.inference.vc/from-instance-noise-to-gradient-regularisation/original ↗

2 Jun 2017

https://www.inference.vc/everything-that-works-works-because-its-bayesian-2/original ↗

1 Jun 2017

https://www.inference.vc/exemplar-cnns-and-information-maximization/original ↗

27 May 2017

https://www.inference.vc/maximum-likelihood-for-representation-learning-2/original ↗

4 May 2017

https://www.inference.vc/unsupervised-learning-by-predicting-noise-an-information-maximization-view-2/original ↗

24 Apr 2017

https://www.inference.vc/evolution-strategies-variational-optimisation-and-natural-es-2/original ↗

6 Apr 2017

https://www.inference.vc/evolutionary-strategies-embarrassingly-parallelizable-optimization/original ↗

5 Apr 2017

https://www.inference.vc/choice-of-recognition-models-in-vaes-a-regularisation-view/original ↗

30 Mar 2017

https://www.inference.vc/comment-on-overcoming-catastrophic-forgetting-in-nns-are-multiple-penalties-needed-2/original ↗

16 Mar 2017

https://www.inference.vc/the-spherical-kernel-divergence/original ↗

9 Mar 2017

https://www.inference.vc/variational-inference-using-implicit-models/original ↗

1 Mar 2017

https://www.inference.vc/variational-renyi-lower-bound/original ↗

11 Feb 2017

https://www.inference.vc/variational-inference-using-implicit-models-part-iii-joint-contrastive-inference-ali-and-bigan/original ↗

9 Feb 2017

https://www.inference.vc/variational-inference-with-implicit-probabilistic-models-part-1-2/original ↗

9 Feb 2017

https://www.inference.vc/variational-inference-with-implicit-models-part-ii-amortised-inference-2/original ↗

9 Feb 2017

https://www.inference.vc/variational-inference-using-implicit-models-part-iv-denoisers-instead-of-discriminators/original ↗

9 Feb 2017

2016

https://www.inference.vc/holiday-special-deriving-the-subpixel-cnn-from-first-principles/original ↗

23 Dec 2016

https://www.inference.vc/my-summary-of-adversarial-training-nips-workshop/original ↗

14 Dec 2016

https://www.inference.vc/solve-intelligence/original ↗

24 Nov 2016

https://www.inference.vc/representation-learning-and-compression-with-the-information-bottleneck/original ↗

11 Nov 2016

https://www.inference.vc/how-to-train-your-generative-models-why-generative-adversarial-networks-work-so-well-2/original ↗

24 Oct 2016

https://www.inference.vc/instance-noise-a-trick-for-stabilising-gan-training/original ↗

21 Oct 2016

https://www.inference.vc/how-powerful-are-graph-convolutions-review-of-kipf-welling-2016-2/original ↗

3 Oct 2016

https://www.inference.vc/are-energy-based-gans-actually-energy-based/original ↗

23 Sept 2016

https://www.inference.vc/infogan-variational-bound-on-mutual-information-twice/original ↗

9 Aug 2016

https://www.inference.vc/temporal-contrastive-learning-for-latent-variable-models/original ↗

14 Jul 2016

https://www.inference.vc/understanding-minibatch-discrimination-in-gans/original ↗

15 Jun 2016

https://www.inference.vc/dilated-convolutions-and-kronecker-factorisation/original ↗

12 May 2016

https://www.inference.vc/notes-on-unsupervised-learning-of-visual-representations-by-solving-jigsaw-puzzles/original ↗

28 Apr 2016

https://www.inference.vc/adversarial-preference-loss/original ↗

25 Mar 2016

https://www.inference.vc/tag/preference-learning/original ↗

25 Mar 2016

https://www.inference.vc/an-alternative-update-rule-for-generative-adversarial-networks/original ↗

24 Mar 2016

https://www.inference.vc/tag/generative-models/original ↗

24 Mar 2016

https://www.inference.vc/tag/adversarial/original ↗

24 Mar 2016

https://www.inference.vc/tag/gan/original ↗

24 Mar 2016

https://www.inference.vc/tag/kl-divergence/original ↗

24 Mar 2016

https://www.inference.vc/deep-learning-is-easy/original ↗

1 Feb 2016

https://www.inference.vc/denoising-as-unsupervised-learning/original ↗

21 Jan 2016

https://www.inference.vc/the-next-episode-2/original ↗

8 Jan 2016