Shao-Hua Sun @shaohua0116 Twitter profile

Pinned Tweet

Shao-Hua Sun

3 months

Our #ICML2024 @icmlconf paper "Diffusion Model-Augmented Behavioral Cloning" led by Shang-Fu Chen and Hsiang-Chun Wang augments BC by learning a diffusion model to capture expert demos and provides gradients to train the policy. Paper: A thread 👉

4

25

151

Last Seen Profiles

@DiamanteCoxinha

@kaede_ub

@Envision0853

@IwasariI

@Gumnam143786

@MizunoVov

@LitPerdida

@MehZahid

@33rdvault

@parcfermeplus

@mr_xtoon

@jandakembangstw

@OdenEnglish

@nongnabenugu

@hothim954

@flyborden

@RAMNIWA16335830

@itxpsp

@Banat_IT

@nvrenlashi

@bokeplokalmalam

@stwmaniax

@miajwes

@dayton_abby

@mitarashi_santa

@bokeplokalmalam

@TheDavidZhou

@amin_snoussi

@zlatansoftic

@bokeplokalmalam

@_AccountsDirect

@EatOffbeat

@brq88206690

@PiFenlon

@bizzlefever_aa

@yamanashiyoukai

Shao-Hua Sun

@shaohua0116

1 year

I am teaching the very first Reinforcement Learning course at National Taiwan University this semester! Dear RL folks, can you describe why students should learn RL in one sentence? Please RT 🙏

64

89

757

Shao-Hua Sun

@shaohua0116

5 years

Review #1 : Weak Reject Review #2 : Weak Reject Review #3 : Reject No one: AC: Ok. I read the reviews and the rebuttal. All the concerns are addressed. ACCEPT. This is an AC that only exists in your wildest dreams. #ICLR2020

Efficient Probabilistic Logic Reasoning with Graph Neural Networks

We employ graph neural networks in the variational EM framework for efficient inference and learning of Markov Logic Networks.

openreview.net

6

122

686

Shao-Hua Sun

@shaohua0116

3 years

I am beyond excited to share that I'll be joining the Electrical Engineering Dept. at National Taiwan University (NTU) as a tenure-track Assistant Professor this fall! I am thrilled for this next chapter and am extremely grateful to everyone who has been part of my Ph.D. journey.

37

10

400

Shao-Hua Sun

@shaohua0116

4 years

I really don't think this discouraging statement should be included for #NeurIPS2020 paper review.

18

21

387

Shao-Hua Sun

@shaohua0116

5 years

#ICLR2020 authors with > 5 submissions: 32 Sergey Levine 20 Yoshua Bengio 16 Cho-jui Hsieh 14 Pieter Abbeel 13 Liwei Wang, Tom Goldstein, Chelsea Finn, Bo Li, Jun Zhu # of accepted papers: 13 Sergey Levine 7 Le Song, Jun Zhu 6 Cho-jui Hsieh, Jimmy Ba, Liwei Wang, Pushmeet Kohli

15

47

348

Shao-Hua Sun

@shaohua0116

3 years

I’m thinking about getting a M1 MacBook Pro. As a person doing research in machine learning, should I be worried about anything? Thanks!

65

19

298

Shao-Hua Sun

@shaohua0116

5 years

The top 5 shortest reviews of #ICLR2020 : 17 words: (weak accept) 27 words (weak accept) 33 words: (weak accept) 33 words: (reject) 40 words: (reject)

Amharic Negation Handling

This work presents Amharic Negation Handling for efficient Sentiment Classification.

openreview.net

13

90

298

Shao-Hua Sun

@shaohua0116

2 months

Trying to book a hotel for #NeurIPS2024 in Vancouver. Every single hotel is so crazily expensive, thanks to Taylor Swift.

16

15

297

Shao-Hua Sun

@shaohua0116

3 years

My #NeurIPS2021 @NeurIPSConf review summary (hours spent reviewing, number of words, score): 11 hrs, 1113 words, 6 15 hrs, 959 words, 4 11 hrs, 1343 words, 7 8 hrs, 841 words, 6 17 hrs, 1571 words, 6 12 hrs, 1277 words, 4 It's like a two-week full-time job (but unpaid).

7

14

228

Shao-Hua Sun

@shaohua0116

1 year

#NeurIPS2023 A reviewer just wrote “thanks for you explanation” to our 600-word rebuttal addressing all the questions he/she asked. Exactly these four words. Just that simple. I’m in shock. Who am I speaking to?

17

8

217

Shao-Hua Sun

@shaohua0116

3 years

I defended my Ph.D. thesis! Words can't describe how thankful I am to my family, advisor @JosephLim_AI , collaborators, dissertation committee members ( @snikolaidis19 , @gauravsukhatme , Prof. Quan Nguyen), and those who joined the defense today.

17

4

207

Shao-Hua Sun

@shaohua0116

2 years

I am proud to present my very first last-author paper as an Assistant Professor at NTU. I'm incredibly fortunate to work with these dedicated and talented students and colleagues: Guan-Ting Liu, En-Pei Hu, Pu-Jen Cheng, @HungyiLee2 A thread 👉

5

31

169

Shao-Hua Sun

@shaohua0116

7 years

Comparisons among SELU (scaled exp. lin. unit), ReLU, and LReLU based on visualization and histogram of activations

3

50

170

Shao-Hua Sun

@shaohua0116

5 years

With a paper accepted by NeurIPS, I finally completed my first machine learning "grand slam": publishing papers at NeurIPS, ICML, and ICLR as a leading author. I am so grateful to all my collaborators as well as the researchers who discuss ideas with me and provide feedback.

11

4

167

Shao-Hua Sun

@shaohua0116

7 years

I implemented Semi-supervised GANs in @tensorflow using generated samples to boost the performance of classifiers

2

49

148

Shao-Hua Sun

@shaohua0116

3 years

1/n This #ICML #ICCV plagiarism issue reminds me of an uncomfortable experience of mine. I believe that the code of my ICLR 2020 paper Program Guided Agent (), which is not publicly released

7

22

143

Shao-Hua Sun

@shaohua0116

5 years

I am thrilled to announce that our NeurIPS 2019 Spotlight paper ”Multimodal Model-Agnostic Meta-Learning via Task-Aware Modulation” is now on arXiv ()!

1

26

129

Shao-Hua Sun

@shaohua0116

3 years

Amazing! #NeurIPS2021 @NeurIPSConf now allows authors to provide feedback to their reviewers on @openreviewnet .

2

6

127

Shao-Hua Sun

@shaohua0116

2 months

#NeurIPS2024 Among the 24 reviewers for our six submissions, 18 responded to our rebuttal (had back and forth discussions with 6), and 13 raised their scores. This is the most engaging rebuttal I have ever experienced. Kudos to our reviewers and ACs!

Shao-Hua Sun

@shaohua0116

2 months

How’s your #NeurIPS2024 rebuttal experience? How much percentage of the reviewers have responded your rebuttal?

0

2

5

122

Shao-Hua Sun

@shaohua0116

3 years

I just passed the qualifying exam. I am now officially a Ph.D. candidate! Special thanks to my qual committee members @JosephLim_AI , @snikolaidis19 , @yanliu_usc , @gauravsukhatme , Prof. Rahul Jain for the valuable feedback and questions to my research!

7

0

118

Shao-Hua Sun

@shaohua0116

5 years

#ICLR2020 submissions have an average score of 3.9067. If your paper got an average score of 5, it beats 81.79% of submissions. Hot keywords are RL, representation learning, generative models, graph neural network, etc.

3

38

104

Shao-Hua Sun

@shaohua0116

5 years

Our #ICLR2020 paper "Program Guided Agent" is accepted as a spotlight paper! We propose a modular framework that can accomplish tasks specified by programs, structured in a formal language, and achieve zero-shot generalization to more complex tasks. Paper:

2

18

99

Shao-Hua Sun

@shaohua0116

3 years

If you are interested in how programs can play a role in improving the generalization of machine learning methods, check out this list of papers related to program synthesis, program induction, program execution, program repair, and programmatic RL.

GitHub - shaohua0116/awesome-program: A curated list of papers related to program synthesis,...

A curated list of papers related to program synthesis, program induction, program execution, program and code repair, and programmatic reinforcement learning. - shaohua0116/awesome-program

github.com

4

24

98

Shao-Hua Sun

@shaohua0116

1 year

I crowdsourced a one-sentence motivation for learning RL on Twitter to kick off the RL course I am teaching at National Taiwan University. I appreciate all the replies! Here's a thread 🧵 organizing the answers I got.

Shao-Hua Sun

@shaohua0116

1 year

I am teaching the very first Reinforcement Learning course at National Taiwan University this semester! Dear RL folks, can you describe why students should learn RL in one sentence? Please RT 🙏

64

89

757

1

8

95

Shao-Hua Sun

@shaohua0116

10 months

#ICLR2024 @iclr_conf score statistics (7304 papers): mean: 5.10; max: 8.67; min: 1.00 >8.5: top 0.59-0.67% 8.0: top 0.68-1.75% 7.5: top 1.93-3.35% 7.0: top 5.20-8.17% 6.75: top 8.42-10.34% 6.5: top 12.13-15.39% 6.25: top 17.48-20.56% 6.0: top 21.02-28.29% 5.75: top 29.07-33.27%

1

8

92

Shao-Hua Sun

@shaohua0116

5 years

As asked by @danijarh and others #ICLR2020 authors with >= 5 submissions sorted by acceptance rate Zhiyuan Li, Mingyuan Zhou, Deva Ramanan 4/5 80.0% Le Song 7/9 77.8% Jimmy Ba 6/8 75.0% Martin Jaggi 5/7 71.4% Abhinav Gupta 5/7 71.4% Pushmeet Kohli 6/9 66.7% Max Welling 5/8 62.5%

4

13

84

Shao-Hua Sun

@shaohua0116

2 years

The first paper from my lab at National Taiwan University w/ my students Hsiang-Chun Wang and Shang-Fu Chen. We employ diffusion models to augment behavioral cloning (BC) and achieve competitive performance in various continuous control domains. A thread👉

2

13

78

Shao-Hua Sun

@shaohua0116

5 months

#ICLR2024 Just arrived in Vienna, Austria 🇦🇹! I will be at ICLR from Monday to Saturday. Drop me a line if you would like to chat! Also, any tourist spots or restaurant recommendations are appreciated!

2

4

77

Shao-Hua Sun

@shaohua0116

23 days

I just booked my #NeurIPS2024 @NeurIPSConf hotel through the official hotel reservation portal. Staying at Pan Pacific Vancouver (closest to the venue) costs $560+/night on Agoda, hotels, etc., and it's only $190 on the portal! Kudos to the organizers.

3

2

76

Shao-Hua Sun

@shaohua0116

10 months

#ICLR2024 @iclr_conf pre/post-rebuttal scores Pre-rebuttal: mean 4.87; max 8.50; min 1.00 Post-rebuttal: mean 5.10; max 8.67; min 1.00 The average score increase this year is probably the lowest compared to all previous ICLRs I have experienced (2019-2023). Why?

8

2

73

Shao-Hua Sun

@shaohua0116

5 years

I just released our #NeurIPS2019 Multimodal Model-Agnostic Meta-Learning (MMAML) code for learning few-shot classification. It contains #PyTorch implementations of our model and two baselines (MAML and Multi-MAML). #NeurIPS #code

1

14

74

Shao-Hua Sun

@shaohua0116

3 years

I am delighted to share that our papers “Learning to Synthesize Programs as Interpretable and Generalizable Policies” () and “Generalizable Imitation Learning from Observation via Inferring Goal Proximity” are accepted by #NeurIPS2021 !

4

5

73

Shao-Hua Sun

@shaohua0116

1 year

Only 1 out of 9 of our #NeurIPS2023 reviewers has responded to our rebuttal. What’s you response rate, my fellow NeurIPS authors?

39

1

72

Shao-Hua Sun

@shaohua0116

5 years

The histogram of #ICLR2020 paper ratings: 1. a paper with ratings 6, 8, 8, got rejected: 2. a paper with ratings 1, 3, 3 got accepted 3. 62% of 3, 6, 8 papers got accepted 4. only 27.8% of 3, 6, 6 papers got accepted

5

13

71

Shao-Hua Sun

@shaohua0116

7 years

I implemented Relation Network and Sort-of-CLEVR in @tensorflow proposed by @DeepMindAI please share&visit my GitHub

2

26

70

Shao-Hua Sun

@shaohua0116

2 years

USC Viterbi PhD Hooding Ceremony @CSatUSC @USCViterbi @USC

2

0

68

Shao-Hua Sun

@shaohua0116

4 years

[Important: Timeline change for NeurIPS 2020] Abstract submission deadline: May 27 Paper submission deadline: June 3 Author notification date: Sept 25

3

17

66

Shao-Hua Sun

@shaohua0116

2 years

Our first ever lab event: furniture assembly 🪑 & pizza 🍕 party!

1

0

62

Shao-Hua Sun

@shaohua0116

1 year

Look who’s in town 👀 @AndrewYNg discusses opportunities in AI at National Taiwan University!

0

1

55

Shao-Hua Sun

@shaohua0116

1 year

#NeurIPS2023 notification date: 9/22 #ICLR2024 abstract submission deadline: 9/21 So this year ICLR does not take papers rejected by NeurIPS anymore?

5

2

55

Shao-Hua Sun

@shaohua0116

3 years

4/n While this might not be comparable to plagiarism, I thought this could raise people's awareness. Please do not use the code of the papers assigned to you for review for non-reviewing purposes, especially without permission from the authors.

1

4

53

Shao-Hua Sun

@shaohua0116

1 year

For those who are anxiously waiting for the decision on ICML 2023 submissions, it probably won't come out soon. The decision notification time was secretly updated to Apr 24 '23 (AoE) on the ICML website. It was Apr 22, according to the archived webpage

2

7

51

Shao-Hua Sun

@shaohua0116

10 months

#NeurIPS2023 Just arrived in New Orleans! I will be at NeurIPS from Sunday to Saturday. Drop me a line or reply if you would like to chat!

0

2

51

Shao-Hua Sun

@shaohua0116

1 year

Thrilled that this work has been accepted to ICML 2023! See you all in Hawaii!

Shao-Hua Sun

@shaohua0116

2 years

I am proud to present my very first last-author paper as an Assistant Professor at NTU. I'm incredibly fortunate to work with these dedicated and talented students and colleagues: Guan-Ting Liu, En-Pei Hu, Pu-Jen Cheng, @HungyiLee2 A thread 👉

5

31

169

2

6

50

Shao-Hua Sun

@shaohua0116

3 years

This went beyond my expectation. The first author of paper B (Duo Li) just posted on Zhihu (Chinese Quora). Summary: 1. The literature survey was done before paper A posted on arXiv 2. The two papers are coincidentally similar since

Pierre Ablin

@PierreAblin

3 years

Imitation is the sincerest form of flattery 🙌🙌🙌

26

103

773

3

4

50

Shao-Hua Sun

@shaohua0116

3 years

Again, this went far beyond my expectation. So many plot twists. Duo Li has completely revised his response to: 1. His ICCV 2021 paper and CVPR 2020 papers plagiarize other papers 2. His co-authors were not involved in plagiarism 3. He feels sorry and will learn from it

Shao-Hua Sun

@shaohua0116

3 years

This went beyond my expectation. The first author of paper B (Duo Li) just posted on Zhihu (Chinese Quora). Summary: 1. The literature survey was done before paper A posted on arXiv 2. The two papers are coincidentally similar since

3

4

50

1

50

Shao-Hua Sun

@shaohua0116

5 months

The first poster session at ICLR 2024. Let’s roll! @iclr_conf

2

5

47

Shao-Hua Sun

@shaohua0116

7 years

I implemented a novel generative model Generative Latent Optimization (GLO) in @tensorflow proposed by FAIR

0

27

45

Shao-Hua Sun

@shaohua0116

1 year

My very first NLP paper got accepted to EMNLP 2023! This is joint work with @lunweiku @windx0303 and the leading student author Nicholas Collin Suwono, which explores the Location-aware Visual Question Generation task, aiming to ask engaging questions from a specific location.

1

2

46

Shao-Hua Sun

@shaohua0116

5 years

I posted my #ICLR2020 rebuttal around two days ago and two of my reviewers have already replied with constructive comments. I’m extremely grateful. I replied to all the rebuttals of the papers that I am reviewing within 6 hours after the rebuttals were posted.

0

45

Shao-Hua Sun

@shaohua0116

5 years

#ICCV2019 2019 stats Top keywords from titles: Learning, Network, Image, Object, Detection, Deep, 3D, Video, Segmentation, Neural Top authors: Ling Shao, Yi Yang, Qi Tian, Luc Van Gool, Xiaogang Wang, Larry S. Davis, Wanli Ouyang, Huchuan Lu, Dacheng Tao, Junjie Yan @ICCV19

1

16

44

Shao-Hua Sun

@shaohua0116

5 years

When will the #ICLR reviews come out? I am so anxious. I even wrote a script that checks my submission every minute and will send me an email once reviews come out.

1

3

43

Shao-Hua Sun

@shaohua0116

2 months

#ICLR2025 enforces authors to review. I’m wondering what would happen if none of the authors has published at ICLR/ICML/NeurIPS before.

6

4

42

Shao-Hua Sun

@shaohua0116

6 months

This is how the privileged stay privileged I guess.

Gautam Kamath

@thegautamkamath

6 months

NeurIPS 2024 will have a track for papers from high schoolers.

78

90

598

1

2

40

Shao-Hua Sun

@shaohua0116

3 years

We are extremely lucky to get reviewers who are responsible and kind - all of our reviewers have responded or updated the ratings. The ratings of our two #NeurIPS2021 submissions before/after rebuttal: 87776665/87777776.

4

0

39

Shao-Hua Sun

@shaohua0116

5 years

The top 5 longest reviews of #ICLR2020 : 3369 words: (weak reject) 2827 words: (reject) 2697 words: (reject) 2298 words: (weak reject) 2290 words: (weak reject)

Winning the Lottery with Continuous Sparsification

We propose a new algorithm that quickly finds winning tickets in neural networks.

openreview.net

2

4

37

Shao-Hua Sun

@shaohua0116

1 year

I talked about "Learning to Synthesize Programs as Interpretable and Generalizable Reinforcement Learning Policies" hosted by @hsuantienlin at TAAI AI Forum 2023, covering our works that re-formulate solving reinforcement Learning tasks as program synthesis problems. 🧵

1

7

38

Shao-Hua Sun

@shaohua0116

5 years

My Gmail app just found a hotel reservation in Hawaii from today’s arXiv update. Apparently, 1. Everyone is going to Hawaii for their holidays. 2. There's a hotel named Dong Yang Inn in Hawaii. 3. There's a paper author named Dong Yang from today's arXiv papers. Thanks, @GoogleAI

3

0

38

Shao-Hua Sun

@shaohua0116

1 year

Pretty much a must for most PhD programs in ML in the States

The PhD Place

@ThePhDPlace

1 year

Are we REALLY expecting people to have published papers BEFORE they apply for a PhD?

68

76

795

3

0

38

Shao-Hua Sun

@shaohua0116

3 years

I would argue that #NeurIPS2019 gives the best conference mug. Change my mind with your marvelous conference mugs!

3

0

35

Shao-Hua Sun

@shaohua0116

3 years

So much stress of PhD students can be attributed to how much power their advisors have. Funding, graduation, recommendation, etc.

2

0

36

Shao-Hua Sun

@shaohua0116

3 months

I will be attending ICML 2024 @icmlconf in Vienna, Austria 🇦🇹! Happy to chat about diffusion models for imitation learning, programmatic RL, PhD in the US ➡️ faculty career in Taiwan, etc. Drop me a line or reply to this tweet if you are interested to connect. Please RT!

3

1

37

Shao-Hua Sun

@shaohua0116

1 year

After 8 months of hard work as an Assistant Professor at National Taiwan University, a ton of effort in researching, teaching courses, mentoring students, and securing government and industry fundings, my lab finally can afford our first-ever robot, a robot lamp.

0

36

Shao-Hua Sun

@shaohua0116

10 months

Since everyone on my Twitter timeline is attending #NeurIPS2023 . I thought it would be helpful to share this Inigo Montoya conference networking tip for initiating conversations, which works great for me.

1

5

35

Shao-Hua Sun

@shaohua0116

5 months

Po-Chen Ko @PoChen23602 , an undergrad from our lab, who will apply for Fall 2025 PhD programs, will present our #ICLR2024 @iclr_conf work. If you are interested in learning from videos or are hiring PhD students, come talk to him at #240 poster in the morning session on May 10.

1

5

35

Shao-Hua Sun

@shaohua0116

7 years

@ImAnishShah @kevin_zakka Here're my comparisons among SELU, ReLU, and LReLU based on visualization and histogram of activations

GitHub - shaohua0116/Activation-Visualization-Histogram: Compare SELUs (scaled exponential linear...

Compare SELUs (scaled exponential linear units) with other activations on MNIST, CIFAR10, etc. - shaohua0116/Activation-Visualization-Histogram

github.com

0

7

34

Shao-Hua Sun

@shaohua0116

5 years

This thread is outrageous. My advisor Prof Joseph Lim let me stay in Taiwan (even after I have been stuck here for two months waiting for my visa) and arranges all the lab meetings according to the time difference between Taiwan and Los Angeles.

1

0

34

Shao-Hua Sun

@shaohua0116

5 months

Jung-Chun Liu @JungChun1013 , a student from National Taiwan University and an incoming @UMich PhD student, will present our #ICLR2024 @iclr_conf work combining the strengths of deep RL and classical planning. Check out our poster in the afternoon session (4:30-6:30 PM) at #270 !

1

2

34

Shao-Hua Sun

@shaohua0116

5 years

Attending @iclr2019 #ICLR2019 ? Come to our poster at 11am on Wed: Composing Complex Skills by Learning Transition Polices. Our use previously acquired primitive skills to compose complex skills (obstacle course, tennis serve, etc.) by learning to smoothly transition between them.

1

7

34

Shao-Hua Sun

@shaohua0116

3 years

I like the title of this #ICLR2022 submission

4

0

32

Shao-Hua Sun

@shaohua0116

5 years

After a long, anxious wait, the #ICLR2020 result is finally out. We are so lucky to have all those hard-working reviewers reviewing our paper and the rebuttal.

1

0

32

Shao-Hua Sun

@shaohua0116

2 months

#ICML2024 @icmlconf Navigating through only ~15% posters in one poster session be like: Going through all of them in 1.5 hours is really a cardio workout.

1

32

Shao-Hua Sun

@shaohua0116

3 years

The paper submission deadline for @iclr_conf #ICLR2022 is in 50 hours. Source:

1

31

Shao-Hua Sun

@shaohua0116

6 years

Check out our code & data for #ECCV2018 paper 𝗠𝘂𝗹𝘁𝗶-𝘃𝗶𝗲𝘄 𝘁𝗼 𝗡𝗼𝘃𝗲𝗹 𝘃𝗶𝗲𝘄: 𝗦𝘆𝗻𝘁𝗵𝗲𝘀𝗶𝘇𝗶𝗻𝗴 𝗡𝗼𝘃𝗲𝗹 𝗩𝗶𝗲𝘄𝘀 𝘄𝗶𝘁𝗵 𝗦𝗲𝗹𝗳-𝗟𝗲𝗮𝗿𝗻𝗲𝗱 𝗖𝗼𝗻𝗳𝗶𝗱𝗲𝗻𝗰𝗲 - synthesize novel target views from multiple source views

0

9

32

Shao-Hua Sun

@shaohua0116

3 years

It's time to panic. #NeurIPS2021

2

5

31

Shao-Hua Sun

@shaohua0116

5 months

I will be presenting our work improving generative adversarial imitation learning (GAIL) by incorporating a diffusion model as a discriminator at the Generative Models for Decision Making workshop at #ICLR2024 . Stop by our poster at 3 PM @ Lehar 3!

0

3

31

Shao-Hua Sun

@shaohua0116

5 years

We are very grateful to our #ICML reviewers who rejected our paper but provided constructive feedback, which greatly contributes to our #NeurIPS paper. We sincerely hope that they see this message. Our ICML title was Model-Agnostic Meta-Learning for Multimodal Task Distributions.

0

4

30

Shao-Hua Sun

@shaohua0116

5 years

I’m very convinced by the @Apple party at #NeurIPS2019 . Whatever the research team at Apple does, sign me up please.

2

0

30

Shao-Hua Sun

@shaohua0116

5 years

The current cumulative sum plot of #ICLR2020 submission scores is as follows. The average scores of 290 submissions have increased. The top 8 successful rebuttals are as follows (with an average score increase ranging from +4 to +2.33).

1

2

29

Shao-Hua Sun

@shaohua0116

5 years

I reviewed three #ICLR2020 submissions. For each paper, I spent roughly 6 hours reading the paper, 2 hours writing the review, and another hour reading the paper once again to make sure the points that I made in the review are valid. My reviews are 900-word long in average.

1

3

29

Shao-Hua Sun

@shaohua0116

5 years

Many of the short reviews simply wrote something like “I don’t know this field at all.” I wonder why they couldn’t talk to ACs to get the paper re-assigned to other reviewers. Was it possible at all for this year ICLR?

2

30

Shao-Hua Sun

@shaohua0116

2 months

Thrilled to share our #COLM2024 @COLM_conf paper "LLM Discussion: Enhancing the Creativity of Large Language Models via Discussion Framework and Role-Play"! This work explores enhancing LLM creativity via multi-LLM discussion and role-play! A thread 👉

2

28

Shao-Hua Sun

@shaohua0116

6 years

Our ICLR2019 paper Composing Complex Skills by Learning Transition Policies is released! We propose transition policies which enable agents to compose complex skills by smoothly connecting previously acquired primitive skills. See our demos and code at

0

8

28

Shao-Hua Sun

@shaohua0116

1 year

I will be attending @icml ! If you are interested in discussing research related to programmatic RL, program synthesis, robot learning, or just wanna meetup, my DM is open! I will be presenting two papers 🧵

3

1

28

Shao-Hua Sun

@shaohua0116

5 years

For those who are traveling to Korea for #ICCV2019 @ICCV19 : as far as I know, *Google map does not work in Korea* and the largest map provider (Naver) does not have an English version. Instead, you should download Kakao map:

Leading you to all the right places, Kakao Map

#kakaobus #kakaomap #kakaometro #map #mobility

www.kakaocorp.com

6

5

27

Shao-Hua Sun

@shaohua0116

5 years

We presented our #NeurIPS2019 spotlight paper: Multimodal Model-Agnostic Meta-Learning via Task-Aware Modulation. Thanks for everyone stopping by! Paper: Code:

1

2

27

Shao-Hua Sun

@shaohua0116

3 years

Hardly seen undergrads without 1st-author paper getting into top labs doing deep learning these days tho. Why is this the case in our field?

Dr. Haley Wahl

@hwahl16

3 years

Please stop telling undergraduates they need a first-author paper to get into grad school. It’s not true and puts way too much pressure and worry on them.

84

262

4K

1

0

26

Shao-Hua Sun

@shaohua0116

7 years

I implemented Representation Learning by Learning to Count that self-supervisedly learns to count visual primitives.

0

13

26

Shao-Hua Sun

@shaohua0116

5 years

My progress in PhD so far.

1

0

25

Shao-Hua Sun

@shaohua0116

3 years

I always feel so sad when the papers that I am reviewing are withdrawn instantly after reviews come out. No discussion. No clarification. No answer to my questions. Nope.

3

0

25

Shao-Hua Sun

@shaohua0116

5 years

My #ICLR2020 preparation today: - Got four vaccinations at once for yellow fever, hepatitis A, typhoid, and polio - Got the medicine for malaria - Registered at

1

3

25

Shao-Hua Sun

@shaohua0116

4 years

Finally met my advisor @JosephLim_AI and lab mate @YoungwoonLee at #ICLR2020 @iclr_conf . It's way too crowded to find each other.

0

1

25

Shao-Hua Sun

@shaohua0116

5 years

So excited to enjoy #NeurIPS2019 in this beautiful city. I will be staying for the whole conference. Let me know if you would like to chat!

3

0

23

Shao-Hua Sun

@shaohua0116

7 years

@hardmaru Looks like SELUs outperform ReLU and LReLU on the datasets I tried (MNIST, Cifar-10, SVHN)

0

7

24

Shao-Hua Sun

@shaohua0116

3 years

Dear #NeurIPS2021 reviewers, Thank you for giving us a break since we posted the rebuttal. We are now mentally prepared for seeing your response. Yours, Authors

1

0

24

Shao-Hua Sun

@shaohua0116

1 year

We will present our #ICML paper today at 2pm ( #116 ), which re-formulates solving an RL task as composing a task-solving program that can be executed to interact with the environment and maximize the return. If you are interested in RL or program synthesis, check out our poster!

0

1

24

Shao-Hua Sun

@shaohua0116

3 years

3/n When I first saw that paper, I did not even know that crosses a line let alone doing something about it, but I definitely felt really uncomfortable. It has been a while and I still have not heard from the authors.

1

0

24

Shao-Hua Sun

@shaohua0116

3 years

5/n Special thanks to @frederick0329 who supports me and encourages me to share this experience.

0

22