{"about":{"non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","site":"https://codewithpapers.app","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page","syntology":{"site":"https://syntology.ai","developers":"https://syntology.ai/developers","mcp":{"server":"https://syntology.ai/mcp","transport":"streamable-http","server_card":"https://syntology.ai/.well-known/mcp/server-card.json","auth":{"type":"trial token, no account","trial_token":"https://syntology.ai/api/oauth/trial/token","method":"POST","docs":"https://syntology.ai/developers"}},"have":"https://syntology.ai/api/graph/have?x=<method, arXiv id or title> (free, answers coverage only)","paper_base":"https://syntology.ai/paper/","atlas_base":"https://app.syntology.ai/?focus="},"machine_readable":[{"url":"https://codewithpapers.app/llms.txt","what":"the machine catalog: every machine-readable file, counted"},{"url":"https://codewithpapers.app/index/manifest.json","what":"paper-to-code index by arXiv id, with Syntology's counts"},{"url":"https://codewithpapers.app/search/manifest.json","what":"site search index (titles, authors) and its files"},{"url":"https://codewithpapers.app/download","what":"bulk files: Syntology's layer, described there"},{"url":"https://codewithpapers.app/build_manifest.json","what":"the build record: inputs, counts, exclusions, probes"}]},"url":"/paper/synthesizing-programs-for-images-using","title":"Synthesizing Programs for Images using Reinforced Adversarial Learning","arxiv_id":"1804.01118","date":"2018-04-03","proceeding":"ICML 2018 7","authors":["Yaroslav Ganin","tejas kulkarni","Igor Babuschkin","S. M. Ali Eslami","Oriol Vinyals"],"abstract":"Advances in deep generative networks have led to impressive results in recent\nyears. Nevertheless, such models can often waste their capacity on the minutiae\nof datasets, presumably due to weak inductive biases in their decoders. This is\nwhere graphics engines may come in handy since they abstract away low-level\ndetails and represent images as high-level programs. Current methods that\ncombine deep learning and renderers are limited by hand-crafted likelihood or\ndistance functions, a need for large amounts of supervision, or difficulties in\nscaling their inference algorithms to richer datasets. To mitigate these\nissues, we present SPIRAL, an adversarially trained agent that generates a\nprogram which is executed by a graphics engine to interpret and sample images.\nThe goal of this agent is to fool a discriminator network that distinguishes\nbetween real and rendered data, trained with a distributed reinforcement\nlearning setup without any supervision. A surprising finding is that using the\ndiscriminator's output as a reward signal is the key to allow the agent to make\nmeaningful progress at matching the desired output rendering. To the best of\nour knowledge, this is the first demonstration of an end-to-end, unsupervised\nand adversarial inverse graphics agent on challenging real world (MNIST,\nOmniglot, CelebA) and synthetic 3D datasets.","url_abs":"http://arxiv.org/abs/1804.01118v1","url_pdf":"http://arxiv.org/pdf/1804.01118v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"synthesizing-programs-for-images-using","repo_url":"https://github.com/wu375/spiral-pytorch-ray","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"synthesizing-programs-for-images-using","repo_url":"https://github.com/wu375/spiral_pytorch_ray","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"reinforcement-learning","task_name":"Reinforcement Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1804.01118","atlas_url":"https://app.syntology.ai/?focus=1804.01118","mcp":null,"developers":"https://syntology.ai/developers","held":true,"reason":"arxiv_id_in_atlas","checked_against":{"atlas_ids":549449,"atlas_ids_sha256":"afd72399dafd9906e9062927f70655952385a3d75488558e1f0484e950ab09b7","read_at":"2026-10-05T01:46:21Z"},"agent_calls":[{"tool":"get_citation_path","arguments":{"paper_1":"1804.01118"},"arguments_in":{"paper_2":"another paper's arXiv id or title"},"call":"get_citation_path(paper_1=\"1804.01118\", paper_2=\"…\")"},{"tool":"get_concepts_for_paper","arguments":{"arxiv_id":"1804.01118"},"call":"get_concepts_for_paper(arxiv_id=\"1804.01118\")"}],"by_repo":[{"repo":"wu375/spiral-pytorch-ray","kind":"listed","samples":0,"ran":0,"instrument":0,"unverified":0},{"repo":"wu375/spiral_pytorch_ray","kind":"listed","samples":0,"ran":0,"instrument":0,"unverified":0}],"by_repo_is":"one row per repository in the archive's code links, Syntology's graph links or the samples; samples 0 means no sample is linked to this paper, a statement about Syntology's coverage, not about the repository: with harvested_for_other_papers present, Syntology harvested the repository for that many other papers, and without it, Syntology harvested nothing from it; placed_by_identical_code counts samples whose link records no repository, placed here because identical code was harvested from this repository for another paper; ran counts executions on a synthesized input and instrument counts failures of Syntology's instrument, not of the code"},"code_links_note":null,"arxiv_metadata":null,"syntology_extracted_results":null,"record_sha256":"3426f1b8be91de786763bc97528b3cb2da2623c2444e1980a3f2f5b12d1cc052","record_changed_at":"2026-09-28","record_changed_at_basis":"first_hashed"}