Sanka Mohottala

2.2K posts

Sanka Mohottala banner
Sanka Mohottala

Sanka Mohottala

@Danny__SM

Graduate Research Assistant at SLIIT || DL Theory, GNN, Human Action Recognition, Network Science || British Humor || Rowing || Humanist

Sri Lanka Katılım Kasım 2013
2.1K Takip Edilen157 Takipçiler
Sanka Mohottala retweetledi
K.
K.@kafkaex·
No one knows you. No one has a story about who you are. No one is waiting for you to be the person you were yesterday. You're just a stranger in a chair by the window, watching a city that doesn't need anything from you. It's the feeling that anything could happen. That the world is bigger than the walls you built around yourself back home. That the life you've been living is just one version of a life, and there are others, and they're not as far away as you thought. At home, you're fixed. Known. You fit into a shape that other people recognize, and after a while, you forget you're even in a shape at all. But here, alone, somewhere new, the shape dissolves. You could be anyone. You could be more of yourself than you've ever been. No one is watching to see if you stay consistent.
Mads@MadsPosting

The spiritual moment when you go out to eat alone at night in another country

English
369
19.9K
117.2K
4.9M
Sanka Mohottala retweetledi
Prof. Anima Anandkumar
Prof. Anima Anandkumar@AnimaAnandkumar·
Solving Inverse PDEs with 1% Paired Data: Introducing Decoupled Diffusion Inverse Solver We propose a data-efficient and physics-aware diffusion framework for solving inverse problems on function spaces. In scientific machine learning, solving inverse problems requires costly and limited data acquisition from physical systems. Existing joint-embedding diffusion models require massive paired training data, as they represent the underlying physics implicitly through statistical correlations. In this work, we identify that under data scarcity, the observation-induced guidance signal vanishes during posterior sampling, making reconstruction impossible. Our Solution: We propose a decoupled design against joint-embedding: an unconditional diffusion learns the coefficient prior, while a neural operator explicitly models the forward PDE for guidance. This enables (1) superior data efficiency (2) effective physics-informed learning and sampling. Performance: Achieves state-of-the-art results on Navier-Stokes, Helmholtz, and Poisson benchmarks, improving spectral error by 54% on average. Data Efficiency: DDIS maintains high accuracy even when limited to just 1% of paired training data, outperforming joint models by 40% in L2 error. Robustness: Theoretical guarantees that avoid the guidance attenuation identified in joint-embedding methods. Check out the paper for the full theoretical analysis and experiments! arxiv.org/abs/2601.23280 Thomas Lin , @jiacheny7, Alex Chiang, Julius Berner, #MachineLearning #DiffusionModels #InverseProblems #PDE #NeuralOperators @Caltech #AI4Science
English
8
38
233
33.1K
Sanka Mohottala retweetledi
PNASNews
PNASNews@PNASNews·
One of the most-viewed PNAS articles in the last week is “Quantifying the compressibility of the human brain.” Explore the article here: ow.ly/3TPc50Y6heR For more trending articles, visit ow.ly/AziX50Y6heS.
PNASNews tweet media
English
4
38
242
48.2K
Sanka Mohottala retweetledi
ₕₐₘₚₜₒₙ
ₕₐₘₚₜₒₙ@hamptonism·
When we turn 18, we wish we took on sports earlier. When we turn 25, we wish we tried harder at school. When we turn 35, we wish we started that business. When we turn 40, we wish to do it all over again. The truth is the best time to plant a seed was 20 years ago. The second best time, is today.
English
65
441
6.8K
643.8K
Sanka Mohottala retweetledi
Ken Ono
Ken Ono@KenOno691·
1/ ANNOUNCING 🎬 MARYAM: The Mirror and the Map, a feature film about Fields Medalist Maryam Mirzakhani (the first woman to win the Fields Medal). After The Man Who Knew Infinity, writer/director Matt Brown, Manjul Bhargava & I are reuniting as associate producers.
Ken Ono tweet media
English
36
287
1.2K
205.1K
Sanka Mohottala retweetledi
Michael Elabd
Michael Elabd@MichaelElabd·
Amazing continual learning paper out of DeepMind 🚨 Most Continual Learning work assumes the backbone is fixed and the burden on the algorithm to fight catastrophic forgetting. This paper flips that assumption on its head and shows pretty convincingly that architecture choices matter just as much for the plasticity–stability trade-off. A few takeaways that stood out to me: Learning vs. retention is heavily architecture-dependent. ResNets and WideResNets are great at picking up new tasks, but they forget aggressively. On the other hand, simple CNNs and even ViTs are surprisingly good at retaining old knowledge, even if they learn new tasks more slowly. Width beats depth. Making networks wider consistently reduces forgetting and improves average accuracy. Making them deeper often gives diminishing returns on learning while worsening forgetting. Pooling is a hidden culprit: Global Average Pooling is a major driver of forgetting because it bottlenecks the final representation. Removing GAP or replacing it with smaller pooling layers significantly improves retention. BatchNorm isn’t always helpful. It helps when task distributions are similar, but under large distribution shifts, BN can accelerate forgetting. What I’d love to see next is this line of work pushed into the LLM regime (larger models, longer task sequences) so we can (1) benchmark continual learning methods more rigorously and (2) start designing architectures explicitly for continual learning, rather than inheriting them from static training.
Michael Elabd tweet media
English
41
188
1.1K
83.3K
The Kobeissi Letter
The Kobeissi Letter@KobeissiLetter·
BREAKING: President Trump announces his new "Board of Peace" which includes: 1. Marco Rubio: US Secretary of State 2. Jared Kushner: Former Advisor to Trump 3. Steve Witkoff: US Special Envoy 4. Tony Blair: Former Prime Minister of the UK 5. Marc Rowan: CEO of Apollo 6. Ajay Banga: World Bank President 7. Robert Gabriel: US National Security Advisor
The Kobeissi Letter@KobeissiLetter

BREAKING: President Trump says he has formed the “Board of Peace” which will be announced soon. Trump says this is the “most prestigious board ever assembled.”

English
865
827
5.1K
1.7M
Sanka Mohottala retweetledi
James Evans
James Evans@profjamesevans·
Thrilled to share our Nature paper, out today, on how AI use has shaped scientific careers and science as a whole (in collaboration with the amazing Qianyue Hao, @xu_fengli, and Li Yong from Tsinghua University and Zhongguancun Academy.) We analyzed tens of millions of research papers spanning four decades of natural science to understand how AI is reshaping science. The findings reveal a paradox. For individual scientists, AI is a career accelerator. Researchers who adopt AI publish 3 times more papers with fewer authors, receive 5 times more citations, and become research leaders more than a year earlier than peers who don't. AI papers appear ~20% more frequently in top-quartile journals. Annual citations run 100% higher than non-AI papers across three decades of follow-up. For science as a whole, AI is a narrowing force. AI-augmented research covers ~5% less topical ground and generates a quarter less engagement among follow-on researchers. The contraction appears in the vast majority of the 200+ subfields we examined. Citation patterns show a starker concentration: in AI research, just 22% of papers capture 80% of all citations. The mechanism is straightforward: AI use has shifted to where data is abundant, at an accelerating pace as models have grown larger. AI gravitates toward well-lit problems and away from foundational and emergent questions where data is necessarily sparse. The result is collective hill-climbing—everyone scaling the same popular peaks rather than searching for higher mountains. This creates "lonely crowds" in the scientific literature: clusters of researchers converging on identical problems without building on each other's work. The pattern holds across biology, chemistry, physics, medicine, materials science, and geology. It persists and increases through each wave of AI—from conventional machine learning through deep learning to today's generative models and LLMs. This isn't inevitable. Models that are powerful at prediction can be inverted to identify what is surprising (to those predictions), and enable us to consider and theorize entailments to surprising new data and findings. But without deliberate intervention, local incentives have and will likely continue to push scientists to optimize and compress what's already known rather than discover what isn't. The history of major discoveries is linked to new ways of seeing nature. If we want AI to accelerate breakthroughs rather than automate the familiar, we need AI systems tuned to surprise that expand sensory and experimental capacity—not just cognition. Paper: rdcu.be/eY5f7 Science commentary: science.org/content/articl… Nature commentary: nature.com/articles/d4158… Nature podcast: nature.com/articles/d4158…
James Evans tweet media
English
10
71
236
252.7K
Sanka Mohottala retweetledi
Yi Ma
Yi Ma@YiMaTweets·
I believe we were the first to formally state the two principles behind intelligence: Parsimony and Self-consistency, in this 2022 position paper: arxiv.org/abs/2207.04630 with Doris Tsao @doristsao and Harry Shum @harryshum For rigorous technical justifications, read our new open textbook: ma-lab-berkeley.github.io/deep-represent… BTW, version 2.0 of the book is coming out soon.
Haider.@slow_developer

Geoffrey Hinton says LLMs are moving beyond imitation toward self-consistent reasoning Instead of just predicting the next word, new models are beginning to identify contradictions in their own logic This unbounded self-improvement will "end up making it much smarter than us"

English
5
24
224
32.9K
Sanka Mohottala retweetledi
Lee Smart
Lee Smart@VFD_org·
This paper quietly crosses a line most frameworks still avoid. It doesn’t just calculate information, it shows that information obeys a continuity law, with local density, flux, and sources/sinks, exactly like energy. That changes the story. What the data actually shows is that precision is not a property of the observer or the detector. It’s a property of the field geometry, specifically, where the system is structurally sensitive to a parameter. In other words: you don’t “measure” truth by probing harder, truth appears where the field can deform with respect to what you’re trying to know. The most important (and least stated) result is this: Energy flow and information flow are not the same thing, and they can strongly decouple. The system can transmit most of its energy forward, while nearly all usable information about its internal state exits through a different channel (even reflection). So a system can look active, efficient, and functional, while becoming informationally opaque. That single observation reframes a lot: • why biological systems fail before they run out of energy • why brains can stay active while losing coherence • why learning systems misalign • why “more data” doesn’t guarantee more understanding Health, stability, and truth all depend on keeping energy transport and information transport phase-aligned. The geometric patterns in the figures aren’t incidental. They are the spatial signature of sensitivity, the legibility map of the system. They show where information is generated, not where power flows. Once you see this, the same structure appears everywhere: mitochondria, neural fields, learning dynamics, sensing systems, even AI. This work doesn’t just improve measurement. It quietly tells us what kind of systems can be known at all. @msahsorin @NaturePhysics @drmichaellevin @KarlFristonNews @SaraWalkerPhD @MillerLabMIT @FrancoisChollet @chaotician
Lee Smart tweet media
Nirosha J. Murugan@niroshajmurugan

This @Nature study shows why energy and information must be understood together. Biology does this too: mitochondria manage energetic flux, while biochemical and electrical signals encode information about stress, state, and repair. Health depends on keeping these two modes aligned: 1) efficient energy distribution and 2) accurate information about perturbations. nature.com/articles/s4156…

English
22
42
241
17.4K
Sanka Mohottala retweetledi
John Carmack
John Carmack@ID_AA_Carmack·
#PaperADay 2 2026: Deep Delta Learning arxiv.org/abs/2601.00417 The standard residual network blocks are limited to adding on top of the existing state, which limits the expressivity of each layer. It is still a universal approximator, but we can always hope for function blocks that are more parameter / performance / training efficient. This paper proposes a new block based on generalizing the Householder matrix so that the state can be partially or completely collapsed onto (or past) a hyperplane in addition to having new values added on top. This allows information along a vector to be multiplicatively erased while a different vector is added. This is an all-math, no-experiments paper, which usually doesn’t bode well. The same scalar gates both the directional collapse and the value addition. Those might be better off independent, like input and forget gates in an LSTM. When the gating vector is 0.0, the layer is an identity. At 1.0, the previous state has been collapsed onto the learnable hyperplane before adding the new value, and at 2.0 the previous state has been reflected across the hyperplane. They force the gate scalar to be bounded between 0 and 2 by using 2*sigmoid, but that means that default initializations will tend to project everything onto the plane, collapsing values quickly, and you can’t actually reach a perfect identity. Is it actually necessary to bound the gate value? Letting it extend past those bounds just results in scaling along the vector, and avoids the sigmoid gradient issues. They normalize the reflection direction after calculating it with an MLP, which can cause some learning dynamics issues since the weight norm can grow without bound, which reduces the effective learning rate with Adam.
English
24
42
466
52.1K
Sanka Mohottala retweetledi
Jorge Bravo Abad
Jorge Bravo Abad@bravo_abad·
Self-wiring neural networks that learn structure without a single gradient Reservoir computing is elegant: feed signals through a random recurrent network, then train only a simple linear readout. It's fast, cheap, and works surprisingly well—until you hit a wall. That static, random wiring was never designed for your task, and performance suffers. Tanguy Cazalets and Joni Dambre take inspiration from neuroscience to fix this. Their method, Hebbian Architecture Generation (HAG), starts from an almost empty reservoir and progressively grows connections between neurons that frequently fire together—embodying the classic maxim "neurons that fire together wire together." No backpropagation, no gradient descent. Just unsupervised structural plasticity guided by long-horizon correlation statistics. The results are striking. On speech and audio classification benchmarks, HAG-grown reservoirs consistently beat traditional Echo State Networks and popular plasticity rules like Intrinsic Plasticity or Anti-Oja learning. On smaller datasets (Japanese Vowels, CatsDogs), HAG matches or exceeds gradient-trained LSTMs and GRUs—while training orders of magnitude faster. The method slashes neuron-to-neuron correlation (from ~0.99 to ~0.47 on Speech Commands), expands effective dimensionality, and dramatically improves class separability metrics like silhouette scores and inter/intra-class distance ratios. What's conceptually satisfying is why it works: by wiring together co-active units over extended time windows, HAG carves out a task-relevant subspace rather than blindly inflating dimensionality. It learns just enough structure to make classes linearly separable—nothing more, nothing less. The broader message: structural self-organization, long studied in biological neural circuits, is a practical route to adaptive machine learning. HAG bridges the efficiency of reservoir computing with the flexibility of learned architectures, all without propagating a single error signal through time. For neuromorphic hardware, physical reservoirs, and low-data regimes, that combination could matter a lot. Paper: nature.com/articles/s4146…
Jorge Bravo Abad tweet media
English
13
52
309
18.9K
Sanka Mohottala retweetledi
Dr. M.F. Khan
Dr. M.F. Khan@Dr_TheHistories·
The painting titled "The Wedding Reception" by artist Tony P Byrne. The painting is part of Byrne's "1960s teen memories" series. It depicts a street scene outside a venue with people gathering for a wedding reception. The artist has mentioned that the lone man walking down the street is the bride's ex-boyfriend. #drthehistories
Dr. M.F. Khan tweet media
English
1
14
159
21.9K
Sanka Mohottala retweetledi
Bratislav Misic
Bratislav Misic@misicbata·
netneurotools: a trainee-oriented approach to network neuroscience | doi.org/10.1101/2025.0… Our lab’s internal toolkit for accomplishing everyday tasks in brain imaging ⤵️
Bratislav Misic tweet media
English
3
86
262
32.7K
Sanka Mohottala retweetledi
Didier 'Dirac's ghost' Gaulin
Didier 'Dirac's ghost' Gaulin@DiracGhost·
Since many of you are interested in the subject of Kolmogorov complexity , I got ya a series of 6 high quality lectures, given by one of the author of the book that I've recommended, Alexander Shen! 👇👇👇👇👇👇👇👇👇👇👇👇🔗
Didier 'Dirac's ghost' Gaulin tweet media
Didier 'Dirac's ghost' Gaulin@DiracGhost

I got two recommendations for you on Kolmogorov complexity, a subject that would be important for mathematicians, computer scientists, but also for philosophers. The first recommendation is for '' Kolmogorov Complexity and Algorithmic Randomness'' by Shen, Uspensky and Vereshchagin (AMS), the second ''An Introduction to Kolmogorov Complexity and its Applications'' by Vitanyi and Li (Springer). Both books are excellent, tons of exercises to practice on, the springer book does go touch on philosophical questions found in probability theory which you will not find in Shen and co. If you are to embark in this journey, i would suggest you to be comfortable with the following preliminaries; theory of computation, basic computational structures, an understanding of big O notation and the basics of probability theory/combinatorics. Have fun!

English
4
32
298
19.1K
Sanka Mohottala retweetledi
Pushmeet Kohli
Pushmeet Kohli@pushmeet·
Happy to announce @GoogleDeepMind's partnership with US Department of Energy (@ENERGY) on the Genesis Mission to transform scientific discovery through AI. We are giving all 17 US National Labs accelerated access to our frontier tools. 🧪 I’m excited to see the AI powered scientific breakthroughs from America’s leading researchers. #AIforScience deepmind.google/blog/google-de…
English
12
47
411
23.2K
Sanka Mohottala retweetledi
Jorge Bravo Abad
Jorge Bravo Abad@bravo_abad·
Compressing higher-order networks without losing what matters Many real systems aren’t just made of pairwise links. A group chat, a coauthored paper, a classroom, or a biochemical complex are group interactions involving 3, 4, or more entities at once. Hypergraphs are the natural way to model this: you put nodes for the entities and “hyperedges” for each group, with one layer for pairs, another for triples, another for quadruples, and so on. The catch: these higher-order models quickly become huge, hard to compute with, and hard to interpret. The key question is: how much of that higher-order structure is really new information, and how much is just redundant with lower orders?  Alec Kirkley, Helcio Felippe and Federico Battiston tackle this with an information-theoretic notion of structural reducibility for hypergraphs. Think of trying to send a whole higher-order network over a very expensive data link. One option is “naïve”: send every layer (pairs, triples, 4-tuples, …) independently. Their alternative is smarter: send only a small set of “representative” layers, then describe the remaining ones as noisy copies of those, using only the differences. The more overlapping structure there is between orders (for example, when all 2- and 3-body interactions are already implied by the 5-body ones), the more you can compress. They turn this into a normalized score η between 0 (no compressibility) and 1 (perfectly nested, fully reducible), and an explicit reduced model that keeps just the non-redundant interaction sizes. Figures in the paper show simple examples where a four-layer hypergraph can be optimally reduced to just two layers while still capturing the essential higher-order organization.  They then stress-test this on synthetic and real data. On controlled “nested” toy hypergraphs, η smoothly decreases as they inject randomness—behaving like a dial from “perfectly structured” to “fully random.” On real systems (coauthorship, contact networks, email threads, tagging systems, etc.), many turn out to be surprisingly compressible: you can drop several hyperedge orders and keep only a small subset of layers, yet preserve global connectivity, community structure, and even the behavior of higher-order voter-model dynamics on top of the network.  The takeaway: you often don’t need the full, unwieldy higher-order description to study a complex system. With the right information-theoretic lens, you can identify which group sizes genuinely add new structure, build a much smaller hypergraph, and still faithfully capture the collective patterns and dynamics you care about.  Paper: journals.aps.org/prl/abstract/1…
Jorge Bravo Abad tweet media
English
11
66
426
33.2K
Sanka Mohottala retweetledi
Jürgen Schmidhuber
Jürgen Schmidhuber@SchmidhuberAI·
Incredibly, even Hinton's recent 2025 article [5] fails to cite Ivakhnenko & Lapa, the fathers of deep learning (1965) [1-3][6-10]. @geoffreyhinton claims [5] that his 1985 "Boltzmann machines" (BMs) [11] (actually 1975 Sherrington-Kirkpatrick models [6]) "are no longer used" but "were historically important" because "in the 1980s, they demonstrated that it was possible to learn appropriate weights for hidden neurons using only locally available information WITHOUT requiring a biologically implausible backward pass." That's ridiculous. This had already been demonstrated 2 decades earlier in the 1960s in Ukraine [1-3]. Ivakhnenko's 1971 paper [3] described a deep learning network with 8 layers and layer-wise training. This depth is comparable to the depth of Hinton's BM-based 2006 "deep belief networks" with layer-wise training [4], published 35 years later without comparison to the original work [1-3] - done when compute was millions of times more expensive. And indeed, over half a century ago, Ivakhnenko's net learned appropriate weights for hidden neurons WITHOUT requiring a biologically implausible backward pass! Hinton & Sejnowski & co-workers have repeatedly plagiarized Ivakhnenko and others, and failed to rectify this in later surveys [6-8]. Crazy fact: today (Fri 5 Dec 2025), the inaugural so-called "Sejnowksi-Hinton Prize" will be handed out at NeurIPS 2025 for a related paper on learning without exact backpropagation [12] which also did not mention the original work on deep learning without backward pass [1-3]. What happened to peer review and scientific honesty? REFERENCES [1] Ivakhnenko, A. G. and Lapa, V. G. (1965). Cybernetic Predicting Devices. CCM Information Corporation. First working Deep Learners with many layers, learning internal representations. [2] Ivakhnenko, Alexey Grigorevich. The group method of data of handling; a rival of the method of stochastic approximation. Soviet Automatic Control 13 (1968): 43-55. [3] Ivakhnenko, A. G. (1971). Polynomial theory of complex systems. IEEE Transactions on Systems, Man and Cybernetics, (4):364-378. [4] G. E. Hinton, R. R. Salakhutdinov. Reducing the dimensionality of data with neural networks. Science, Vol. 313. no. 5786, pp. 504-507, 2006. [5] G. Hinton. Nobel Lecture: Boltzmann machines. Rev. Mod. Phys. 97, 030502, 25 August 2025. [6] J.S. A Nobel Prize for Plagiarism. Technical Report IDSIA-24-24 (2024, updated 2025). [7] J.S. How 3 Turing awardees republished key methods and ideas whose creators they failed to credit. Technical Report IDSIA-23-23, Dec 2023. [8] J.S. (2025). Who invented deep learning? Technical Note IDSIA-16-25. [9] J.S. (2015). Deep learning in neural networks: An overview. Neural Networks, 61, 85-117. Got the first Best Paper Award ever issued by the journal Neural Networks, founded in 1988. [10] J.S. Annotated History of Modern AI and Deep Learning. Technical Report IDSIA-22-22, 2022, arXiv:2212.11279. [11] D. Ackley, G. Hinton, T. Sejnowski (1985). A Learning Algorithm for Boltzmann Machines. Cognitive Science, 9(1):147-169. [12] T. P. Lillicrap, D. Cownden, D. B. Tweed, C. J. Akerman. Random synaptic feedback weights support error backpropagation for deep learning. Nature Communications vol. 7, 13276 (2016).
Jürgen Schmidhuber@SchmidhuberAI

The #NobelPrize in Physics 2024 for Hopfield & Hinton turns out to be a Nobel Prize for plagiarism. They republished methodologies developed in #Ukraine and #Japan by Ivakhnenko and Amari in the 1960s & 1970s, as well as other techniques, without citing the original inventors. None of the important algorithms for modern AI were created by Hopfield & Hinton. Today I am releasing a detailed tech report on this [NOB]: people.idsia.ch/~juergen/physi… Of course, I had it checked by neural network pioneers and AI experts to make sure it was unassailable. Is it now acceptable for me to direct young Ph.D. students to read old papers and rewrite and resubmit them as if they were their own works? Whatever the intention, this award says that, yes, that is perfectly fine. Some people have lost their titles or jobs due to plagiarism, e.g., Harvard's former president [PLAG7]. But after this Nobel Prize, how can advisors now continue to tell their students that they should avoid plagiarism at all costs? It is well known that plagiarism can be either "unintentional" or "intentional or reckless" [PLAG1-6], and the more innocent of the two may very well be partially the case here. But science has a well-established way of dealing with "multiple discovery" and plagiarism - be it unintentional [PLAG1-6][CONN21] or not [FAKE,FAKE2] - based on facts such as time stamps of publications and patents. The deontology of science requires that unintentional plagiarists correct their publications through errata and then credit the original sources properly in the future. The awardees didn't; instead the awardees kept collecting citations for inventions of other researchers [NOB][DLP]. Doesn't this behaviour turn even unintentional plagiarism [PLAG1-6] into an intentional form [FAKE2]? I am really concerned about the message this sends to all these young students out there. REFERENCES [NOB] J. Schmidhuber (2024). A Nobel Prize for Plagiarism. Technical Report IDSIA-24-24. people.idsia.ch/~juergen/physi… [NOB+] Tweet: the #NobelPrize in Physics 2024 for Hopfield & Hinton rewards plagiarism and incorrect attribution in computer science. It's mostly about Amari's "Hopfield network" and the "Boltzmann Machine." x.com/SchmidhuberAI/… (1/7th as popular as the original announcement by the Nobel Foundation) [DLP] J. Schmidhuber (2023). How 3 Turing awardees republished key methods and ideas whose creators they failed to credit. Technical Report IDSIA-23-23, Swiss AI Lab IDSIA, 14 Dec 2023. people.idsia.ch/~juergen/ai-pr… [DLP+] Tweet for [DLP]: x.com/SchmidhuberAI/… [PLAG1] Oxford's guide to types of plagiarism (2021). Quote: "Plagiarism may be intentional or reckless, or unintentional." web.archive.org/web/2021122714… [PLAG2] Jackson State Community College (2022). Unintentional Plagiarism. [PLAG3] R. L. Foster. Avoiding Unintentional Plagiarism. Journal for Specialists in Pediatric Nursing; Hoboken Vol. 12, Iss. 1, 2007. [PLAG4] N. Das. Intentional or unintentional, it is never alright to plagiarize: A note on how Indian universities are advised to handle plagiarism. Perspect Clin Res 9:56-7, 2018. [PLAG5] InfoSci-OnDemand (2023). What is Unintentional Plagiarism? [PLAG6] Copyrighted.com (2022). How to Avoid Accidental and Unintentional Plagiarism (2023). Copy in the Internet Archive. Quote: "May it be accidental or intentional, plagiarism is still plagiarism." [PLAG7] Cornell Review, 2024. Harvard president resigns in plagiarism scandal. 6 January 2024. [FAKE] H. Hopf, A. Krief, G. Mehta, S. A. Matlin. Fake science and the knowledge crisis: ignorance can be fatal. Royal Society Open Science, May 2019. Quote: "Scientists must be willing to speak out when they see false information being presented in social media, traditional print or broadcast press" and "must speak out against false information and fake science in circulation and forcefully contradict public figures who promote it." [FAKE2] L. Stenflo. Intelligent plagiarists are the most dangerous. Nature, vol. 427, p. 777 (Feb 2004). Quote: "What is worse, in my opinion, ..., are cases where scientists rewrite previous findings in different words, purposely hiding the sources of their ideas, and then during subsequent years forcefully claim that they have discovered new phenomena."

English
39
47
347
56.1K
Sanka Mohottala retweetledi
Sergey Levine
Sergey Levine@svlevine·
This turns out to work really well as a way to do curriculum learning. E.g., we first train basic jumping with motion imitation, then finetune to maximize jumping performance with buffer initialization from imitation. Similar idea for hind legs walking
Sergey Levine tweet mediaSergey Levine tweet media
English
1
1
16
2.3K

Keşfet

@jiacheny7 @Caltech @KobeissiLetter @xu_fengli @doristsao @harryshum @NaturePhysics @drmichaellevin